text
stringlengths 18
1.23k
| labels
class label 7
classes | keywords
stringlengths 5
6.15k
|
|---|---|---|
Calculated embeddings for the entire dataset and saved in Amazon S3 | 3Machine Learning Engineer
| pytorch, numpy, lightgbm, rest api, sas, pandas, databases, aws batch, python, github, huggingface, nltk, time series, ml frameworks, aws sdk for python (boto3), dvc, sql, aws eventbridge, aws api gateway, c++, mlflow, aws lambda, aws ec2, go, ibm, tensorflow, xgboost, streamlit, nlp, aws sns, terraform, matlab, pyspark, docker, aws ecs, matplotlib, fastapi, statistics, aws, nginx, aws ecr, aws redshift, scikit-learn, aws s3, aws sdk for python, data, github actions, computer vision |
Developed a predictive behavior model with an accuracy score of 70% to improve customer retention for the seventh-largest bank in Argentina, HSBC. Tools used included SSIS, R, and XGBoost. | 5Analytics Engineer
| apache spark, teradata, c#, jupyter notebooks, rest api, pandas, qlik, python, powerbi, gcp storage, github, gcp, azure data factory, redhat, mongodb, gcp bigquery data, scala, sql, aws api gateway, dynamodb, aws lambda, google dataproc, aws dynamodb, ibm, google cloud storage, sql server, r, great expectations, azure, pyspark, docker, power bi, decision making, mysql, aws, java, azure sql, databricks, jupyter notebook, data management, flask, aws s3 |
Migrated a monolithic application written in Java to a microservices architecture based in Go, AWS, Docker, PostgreSQL and Kubernetes, improving the performance, resilience and maintainability of the systems. | 6Full-Stack Engineer
| other |
Developed exploratory and cohort analysis using Python to discover patterns, test hypotheses and assumptions on the initial market research. | 1Data Analyst
| snowflake, sas, pandas, powerquery, looker studio, dax, sql, looker, python, data, powerbi, visualization, decision making |
Participated in the developing of data pipelines using Spark on AWS EMR and feature engineering for a state-of-the-art recommender system, which was accepted and ranked 6 in the ACM RecSys 2020 challenge. | 0Data Engineer
| aws ec2, trax, apache spark, tensorflow, tableau, sql server, r, google data, aws s3data, nlp, postgresql, jenkins, apache cassandra, python, git, nbgrader, docker, powerbi, github, latex, huggingface, sqlserver, aws emr, aws vpc, aws rds, aws, ml frameworks, c, aws emrvisualization, aws redshift, scikit-learn, apache airflow, flask, scala, aws s3, sql, google data studio, aws mwaa |
Created various applications for audiovideo processing and transcoding using Python and C. | 4Software Engineer
| pytorch, apache spark, c#, oracle, spark, python, git, powerbi, gcp, aws glue, apache airflow, excel, bigquery, sql, singlestore, c++, postgres, tensorflow, aws codepipeline, postgresql, docker, power bi, microsoft excel, matplotlib, linux, plotly, java, aws, aws ecr, databricks, aws redshift, scikit-learn, asp.net, airflow, dash, data, jira, aws kinesis, redshift |
Developed a transactional risk model utilizing SAS Mining and Base SAS, which was successfully deployed across over 20 program allies. The model was deployed using AWS EC2, leveraging its scalability and computational capabilities for efficient implementation. This initiative aimed to enhance risk assessment and mitigation strategies for the program and its allies for a loyalty reward company. | 5Analytics Engineer
| pytorch, numpy, apache spark, jupyter notebooks, tableau, sas, pandas, databases, azure blob storage, javascript, python, powerbi, github, gcp, huggingface, ml frameworks, azure data factory, google compute engine, azure virtual machines, aws sdk for python (boto3), sql, aws api gateway, visualization, mlflow, aws lambda, aws ec2, aws boto3, go, xgboost, azure synapse analytics, sql server, streamlit, bash, nlp, postgresql, terraform, azure, pyspark, azure synapse, docker, matplotlib, mysql, fastapi, plotly, aws, aws ecr, azure sql, databricks, jupyter notebook, data management, scikit-learn, shell scripts, aws s3, data, github actions |
Designed and developed a REST API using FastAPI as the backend of a remote recruiting website with the purpose of connecting recruiters with data professionals. The backend was the core of a complex microservices architecture that coordinated Machine Learning APIs and several data sources such as Amazon S3, PostgreSQL, and MongoDB. | 3Machine Learning Engineer
| pytorch, docker compose, numpy, tableau, rest api, databases, pandas, css, javascript, python, librosa, git, compose, github, tfjs, ml frameworks, html, tfx, node.js, mongodb, tf serving, dvc, tensorflowjs, nodejs, apache airflow, keras, tensorflow serving, sql, c++, mlflow, postgres, aws lambda, aws ec2, tensorflow, r, ownership, nlp, postgresql, scrapy, collaboration, terraform, sqlalchemy, docker, aws ecs, aws vpc, mysql, fastapi, plotly, statistics, aws, numpyvisualization, opencv, autogluon, kotlin, scikit-learn, airflow, aws s3, dash, data, pytest, github actions, computer vision, react |
Created an NLP-based writing assistant that helps users select the most relevant parts of a text with respect to a given description. For this, BERT was used in a zero-shot classification fashion as well as SentenceBERT for sentence embeddings. | 2Data Scientist
| api gateway, numpy, generated automatic visualizations and reports using plotly, performed geospatial analysis on time dependent data using numpy, eafit university, developed a loan default prediction model using xgboost and catboost models, deep learning specialization, aws emr, and geopandas., seaborn, data and deployment specialization, pinecone, keras, langchain, sql, bsc, tensorflow, microsoft word, geospatial analysis, postgresql, data engineer nanodegree, engineering science, research consultant, terraform, matlab, docker, aws ecs, matplotlib, and serving the explained predictions through a dockerized api., fastapi, catboost, golang, applying an explainability framework, consulting analyst, north university oct 2020, aws redshift, scikit-learn, stack overflow questions, s3, which allowed for real-time predictions., github actions, eafit university march 2020, databases, labview, aws ecr, created data-driven models describing the space-time connections of a public water supply system., python, implemented etl pipelines for data ingestion pulling data from row formats and a data lake., github, matplotlib., ml frameworks, analyzed the activity patterns of a nationwide online course and its correlation with students grades., mongodb, using fine tuned distilbert and roberta transformer architectures., apache airflow, qdrant, mechanical engineering, visualization, mlflow, provided feedback on the course content and its impact on students performance., llms, githubdeveloped and deployed an end to end nlp assistant that helps users write better questions in online forums e.g., aws lambda, aws ec2, aws fargate, go, domo, xgboost, transformers, streamlit, bash, nlp, scrapy, created two one-stage object detectors using both the yolov4 model and retinanet fpn. these models were served through a dockerized api deployed in aws ecs, mysql, education, plotly, aws, aws ecr, redis, airflow, aws s3, data, scikit learn, computer vision |
Created some learning virtual objects with HTML, CSS and Javascript in an educational program for the Colombian government.. | 4Software Engineer
| qemu., taught courses in analysis and design of algorithms, numpy, c#, provided support to students in solving problems related to competitive programming., bitbucket, pandas, databases, css, functional programming, javascript, python, vagrant, github, neural networks, ml frameworks, html, sphinx, computer simulation, keras, spring, sql, visualization, c++, django, virtualbox, aws ec2, joomla, go, and fundamentals of programming languages., sql server, bash, postgresql, universidad del valle mar, coupled with my ability to work collaboratively with teams, azure, matplotlib, latex, mysql, linux, java, aws, azure sql, microsoft azure, discrete mathematics, data, managed software projects related to recommendation systems and learning virtual objects., github actions, adjunct professor, enable me to deliver high-quality software solutions according to the client expectations. im committed to staying up-to-date with emerging technologies and trends in the field to ensure i continue to deliver exceptional results.skillsprogramming, php |
Built APIs exposing discrete simulations and statistical models for demand forecasting using Statsmodels and scikit-learn to assist managers in workforce sizing and resource allocation. | 2Data Scientist
| pytorch, apache spark, bitbucket, circleci, snaplogic, spark, javascript, kubernetes, python, powerbi, gcp, time series, angular, node.js, snowflake, aws glue, nodejs, apache airflow, splunk, planning, sql, dynamodb, aws lambda, aws ec2, aws dynamodb, graphql, google bigquery, facebook prophet, aws sns, docker, astronomer, fivetran, decision making, aws, c, typescript, databricks, aws redshift, scikit-learn, dbt, flask, airflow, aws s3, memsql data, redshift, scrum, aws mwaa |
Implemented CICD pipelines to automate infrastructure deployment using Github Actions and Terraform. | 0Data Engineer
| apache spark, databases, css, python, git, powerbi, github, data modeling, aws emr, html, snowflake, delta lake, aws glue, apache airflow, splunk, sql, shell, dynamodb, aws lambda, aws ec2, aws dynamodb, sigma, r, postgresql, aws sns, terraform, azure, matlab, pyspark, docker, latex, azure devops, aws rds, aws, sqlite, aws step functions, azure sql, databricks, aws redshift, dbt, shell scripts, aws s3, data, github actions, aws kinesis, delta live tables |
Partnered with management to deliver ad hoc reporting for inventory management and backorder tracking using Tableau, Alteryx, SAP IBP, and Excel. | 5Analytics Engineer
| sap, tableau, oracle, oracle cloud, dax, postgresql, cloud sql, power bi, microsoft excel, powerbi, mysql, alteryx, snowflake, vba, dbt, excel, planning, sql, data |
Designed and built a data warehouse following Kimball’s data modeling technique and the modular data modeling technique using DBT and SQL, applying tests with DBT over data sources and transformed tables inside the data warehouse. | 0Data Engineer
| snowplow, apache spark, circleci, javascript, kubernetes, heroku, looker, python, git, google bigquerydata, gcp, data modeling, aws athena, gitlab, node.js, snowflake, delta lake, mongodb, aws glue, nodejs, model-view controller, apache airflow, sql, elasticsearch, aws api gateway, dynamodb, visualization, airbyte, aws ec2, aws dynamodb, ibm, great expectations, postgresql, aws sns, elastichsearch, terraform, jinja, azure, pyspark, docker, aws ecs, decision making, azure devops, mysql, aws rds, linux, aws, opencv, azure sql, databricks, aws redshift, dbt, aws s3, google data studio, delta live tables |
Migrated complex pipelines from Matillion using technologies such as Kotlin and Spark. | 0Data Engineer
| apache spark, aws cloudformation, oracle, sql servers, dax, spark, dax., cloud sql, python, git, powerbi, programming, rds, aws emr, cloud, gitlab, matillion, azure data factory, dbschema, apache airflow, big data processing, sql, visualization, aws lambda, sap, azure analysis services., bash, postgresql, azure, docker, azure data warehouse., aws rds, aws, lambda. azure data factory, gradle., azure sql, kotlin, aws redshift, aws s3, kotlin., power bi., redshift, emr, oracle dbs |
Led several data engineering projects, that include the transformation of OLTP + API Unstructured Data to OLAP Schemas deployed in redshift. | 3Machine Learning Engineer
| apache spark.visualization, bokeh, lightgbm, tensorflow, pandas, twisted., postgresql, jenkins, twisted, apache cassandra, python, git, docker, aws ecs, github, amazon redshift, and javascript.databases, fastapi, aws, ml frameworks, mongodb.data, mongodb, aws redshift, scikit-learn, apache airflow, flask, c++, github actions, mlflow, computer vision |
Built ETL workflow using Spark, Databricks, Redshift and Big Query leveraging on UDF to enrich Mlabs data with MaxMind API to periodically update OLAP project data sources to measure ISP’s geo-reference speed. | 0Data Engineer
| cloudformation, pytorch, aws cloudformation, relational dbs, socket.io, mssql and, javascript, shapely, git, powerbi, selenium, aws emr, cloud, azure data factory, node.js, aws glue, glue, qgis, nosql, sql, dynamodb, athena., azure sql datawarehouse, vpcs. microsoft azure., django, socket io, geoanalytics, backend, postgresql, azure devops., azure, azure synapse, docker, power bi, aws ecs, metabase., spectrum, databricks, mssql, aws redshift, aws stack, html., aws sqs, databricks sql, delta lake, apache spark, tableau, databases, pandas, spark, cloud sql, python, programming, html, big data processing, scala, postgis, visualization, metabase, aws lambda, aws ec2, esp. lambda, go, google bigquery, r, flask., h3, azure devops, mysql, aws rds, apache sedona, java, aws, c, azure sql, dbt, ms sql, flask, aws s3, rpa, data, redshift |
Captured business information needs of the commercial and debt collection teams. | 1Data Analyst
| sap, power automate, tableau, microsoft word, pandas, jinja, python, git, sharepoint, microsoft excel, powerbi, github, decision making, sap business objects data services, sap business objects web, vs code, microsoft power point, ms excel, minitab visualization, snowflake, mssql, oracle data, dbt, apache airflow, ms sql, planning, sql, sap business objects bi, matlab databases, power apps, sigma |
Designed and implemented an event-based system to export BigQuery tables to Treasure Data (a consumer data platform) as soon as they were updated using Python, Pyspark, Embulk, Kubernetes, Pubsub, Cloud Functions, Keda and Redis, reducing the synchronization time between platforms, form one day to minutes, thus decreasing the time marketing teams had to wait to start their campaigns once audiences were ready. | 0Data Engineer
| apache spark, kafka, cloudwatch, tableau, apache kafka, databases, pandas, jenkins, hive, kubernetes, looker, python, cloud sql, gcp, data modeling, microsoft access, aws emr, ml frameworks, prestosql, pubsub, vba, seaborn, apache airflow, planning, sql, aws ebs, digdag, visualization, c++, sigma, google dataproc, aws ec2, caret, google cloud storage, xgboost, google bigquery, r, streamlit, bash, postgresql, embulk, terraform, pyspark, aws cloudwatch, docker, matplotlib, sparkml, mysql, aws rds, fastapi, plotly, google cloud, aws, aws cloudfront, rpart, apache hive, dataform, data studio, databricks, jupyter notebook, aws redshift, scikit-learn, dbt, flask, redis, airflow, aws s3, dataproc, data |
Used Web Scraping in python (Requests - Beautiful Soup - Pandas) to get data from web sites that were legally approved to be scraped and usually contain points of interest. This data must be cleaned and standardized to be ready to enrich HERE map content. | 1Data Analyst
| other |
Implemented and deployed an event-driven serverless app in Slack using Python, AWS Lambda, AWS DynamoDB, AWS EventBridge and AWS S3 to automate operational requests across the organization, reducing by ~60% operations processes time. Standardized and centralized in DynamoDB data from multiple sources (Airtable Google Sheets AWS RDS). | 0Data Engineer
| apache spark, tableau, databases, python, powerbi, github, data modeling, nltk, snowflake, aws glue, apache airflow, sql, aws eventbridge, dynamodb, visualization, google sheets, airbyte, aws lambda, aws ec2, aws dynamodb, bash, postgresql, aws sns, terraform, matlab, docker, power bi, decision making, aws rds, aws, aws step functions, airtable, databricks, jupyter notebook, aws redshift, dbt, redash, aws s3, data, github actions, aws kinesis, leadership |
Developed a modular code solution using functional programming supported in python (Pandas, NumPy), reducing time to market and increasing its quality by minimizing human interaction. This solution reduces manual workflow time from 24 working hours to 30 minutes and it is now being applied in Latin America. | 1Data Analyst
| other |
Established CICD pipelines and automated infrastructure deployment in Terraform and GitHub Actions. | 0Data Engineer
| apache spark, tableau, databases, python, powerbi, github, data modeling, nltk, snowflake, aws glue, apache airflow, sql, aws eventbridge, dynamodb, visualization, google sheets, airbyte, aws lambda, aws ec2, aws dynamodb, bash, postgresql, aws sns, terraform, matlab, docker, power bi, decision making, aws rds, aws, aws step functions, airtable, databricks, jupyter notebook, aws redshift, dbt, redash, aws s3, data, github actions, aws kinesis, leadership |
Built data pipelines following event-driven approaches on transactional systems. | 0Data Engineer
| pytorch, scikit, apache spark, python, git, github, data modeling, google identity platform, redcap, aws glue, apache airflow, sql, dynamodb, aws lambda, aws ec2, aws dynamodb, google cloud run, go, google cloud storage, google bigquery, r, ml tools, streamlit, nlp, postgresql, aws sns, terraform, matlab, docker, latex, aws rds, linux, statistics, aws, java, aws step functions, aws redshift, firestore, dbt, aws s3, data, github actions, aws kinesis, leadership, computer vision |
Identified products that were more sensitive for discounts, in order to help the store know which products they could offer to their clients, using price elasticity technique through a Log-Log regression model, indicating which purchases would be profitable to offer at a lower price. | 2Data Scientist
| caret, lightgbm, apache spark, r, databases, swift, nlp, spark, raster, pyspark, microsoft office, looker, python, git, microsoft excel, shiny, matplotlib, decision making, time series, plotly, statistics, ml frameworks, microsoft power point, databricks, vscode, jupyter notebook, seaborn, ggplot2, scikit-learn, bigquery, sql, shap, data, visualization, microsoft powerpoint |
Defining and executing data governance strategies. | 0Data Engineer
| other |
Advised in big data pipeline design for POCs (inhouse & client). | 0Data Engineer
| apache spark, sql server, apache kafka, dataflow, centos, postgresql, jenkins, azure, kubernetes, apache cassandra, python, apache hadoop, cassandra, data modeling, aws emr, mysql, datastax, aws, cql, tidb, azure sql, databricks, dagster, apache airflow, azure sq, apache flink, scala, sql, data, azure cosmosdb, hadoop, metabase |
Created a Near Real-Time transaction, refund and chargeback processing platform to build an analytics platform on top of a data lake using Scala, Python, Spark, Databricks, Hadoop (On-Prem), Airflow, RabbitMQ achieving real-time data visualization for decision making and information management for an electronic payments company. | 0Data Engineer
| oracle, bitbucket, dataflow, circleci, jenkins, javascript, gcp, confluent schema registry, flyway, aws emr, aws athena, aws glue, datadog, planning, testcontainers, bigquery, sql, spring framework, django, sap, ibm, google cloud storage, kinesis, gcp dataflow, postgresql, docker, design patterns, decision making, fastapi, jquery, databricks, sns, kotlin, aws redshift, gcs, oracle db, aws sqs, s3, pytest, github actions, hadoop, aws kinesis, sap erp, aws neptune, junit 5, msk, apache spark, tableau, rest api, apache kafka, databases, pandas, apache cassandra, python, apache hadoop, structured streaming, data engineering, github, rds, apache airflow, spring, scala, sqs, elasticsearch, visualization, spark structured streaming, airbyte, aws ec2, go, aws sns, collaboration, pyspark, cassandra, mysql, aws rds, java, aws, apache beam, dbt, aws msk, kotest, redis, airflow, aws s3, redshift |
Fine-tuned on top layers in LLMs (Large Language Models) in large medical datasets for similarity through encoder-decoder transformers models for conditional generation and self-instruct GPT for prompt-learning through Pytorch, hugging face library and OpenAI API. | 3Machine Learning Engineer
| pytorch, numpy, spacy, aws cloudformation, oracle, dax, rabbitmq, jenkins, ibm db2, kubernetes, powerbi, gcp, huggingface, jupyter, ubuntu, dvc, seaborn, keras, prefect, sql, django, tensorboard, gitlab ci, ibm, tensorflow, dask, centos, postgresql, aws eks, azure, aws cloudwatch, docker, aws ecs, cerberus, matplotlib, decision making, fastapi, swagger, opencv, jupyter notebook, scikit-learn, oracle db, fast.ai, scipy, grafana, github actions, php, cockroachdb, apache spark, kubeflow, databases, pandas, apache hadoop, python, github, sql serveranalysisservices, elyra, gitlab, ml frameworks, mongodb, apache airflow, elasticsearch, visualization, celery, mlflow, aws lambda, aws ec2, db2, go, jupyterhub, sql server, transformers, streamlit, nlp, pyspark, ansible, fastai, jupyterlab, aws, java, azure sql, yellowbrick, flask, redis, airflow, aws s3, data, zabbix, mlxtend, computer vision |
Exploratory multi-dimensional data analysis. Develop algorithms for consensus unsupervised clustering of billions of data. | 2Data Scientist
| other |
Built and maintained scalable data pipeline workflows using Apache Airflow to orchestrate Docker jobs on EKS, and DBT to build a dimensional model on AWS Redshift for an important parking company in Colombia. | 0Data Engineer
| feast, apache spark, tableau, pandas, jenkins, css, javascript, aws batch, python, git, github, gcp, angular, gitlab, html, aws glue, apache airflow, splunk, bigquery, sql, tablea, dynamodb, c++, mlflow, airbyte, aws lambda, google dataproc, aws ec2, aws dynamodb, gitlab ci, express.js, sql server, nlp, great expectations, postgresql, aws sns, terraform, aws eks, matlab, pyspark, docker, microsoft excel, design patterns, visual basic, mysql, aws rds, vs code, linux, statistics, aws, java, aws step functions, databricks, mssql, aws redshift, dbt, ms sql, aws s3, dataproc, data, pytest, github actions, aws kinesis |
Cleaned, extracted and analyzed data from SQL using the Statistics model. | 1Data Analyst
| other |
Designed and managed web applications and websites for a wide range of clients, including high-profile athletes, esteemed chefs, construction firms, advertising agencies, beauty and care product companies, and coupon websites. The technology stack utilized for these projects comprised Ruby on Rails, Sinatra, Wordpress, and Flask, each carefully customized to cater to the distinct requirements of the respective industries and clients. | 4Software Engineer
| numpy, kafka, apache kafka, redux, databases, pandas, perfect, css, javascript, heroku, python, github, gcp, selenium, programming, time series, ml frameworks, html, node.js, software architecture, mongodb, apache airflow, keras, spring, planning, pymongo, ruby on rails, react.js, visualization, postgres, travisci, django, teamwork, bokeh, graphql, go, express.js, tensorflow, xgboost, r, ownership, storybook, streamlit, bash, postgresql, collaboration, pyspark, docker, microsoft excel, sqlserver, mysql, plotly, statistics, aws, nginx, java, c, opencv, jquery, typescript, aws redshift, scikit-learn, sidekiq, flask, airflow, data, pytest, sinatra, ruby, php, react |
Built state-of-the-art recommender system for social network data, which was presented and ranked 6 in the ACM RecSys 2020 challenge. | 3Machine Learning Engineer
| aws ec2, pytorch, typer, numpy, tensorflow, xgboost, postgresql, facebook prophet, prophet, pythondatabases, pyspark, scipyvisualization, python, docker, git, matplotlib, latex, huggingface, mysql, time series, fastapi, jax, aws, ml frameworks, databricks, mongodb, seaborn, scikit-learn, react, flask, apache impala, shap, scipy, mlflow, scikit learn, mongodbdata |
Lead a team that built the analytics department in a large CPG company : o Built an image classifier for a marketing campaign. o Formulate KPI to monitor several company activities. o Applied machine learning to several business use cases. | 2Data Scientist
| statistical modeling, pytorch, teaching, apache spark, databases, julia, spark, pytorch., python, data engineering, github, google, gcp, java and julia., programming, time series, ml frameworks, lambda functions, relational, massive datasets, real time bidding, nosql, visualization, use cases, tensorflow, r, applied machine learning, nlp, programming languages, machine learning, deep learning, machine translation, kpi, latex, nosql dbs, aws rds, aws, java, ai, reinforcement learning, aws stack, scikit-learn, statistical, marketing campaign, data, customer reviews, computer vision |
Development of web apps using Angular - Python stack. | 6Full-Stack Engineer
| db2, ibm, tensorflow, apache spark, ownership, databases, spark, facebook prophet, prophet, kubernetes, microsoft office, python, docker, git, design patterns, time series, angular, fastapi, linux, aws, ml frameworks, node.js, databricks, aws redshift, aws glue, flask, openai embeddings service, data, visualization |
Collaborated in the implementation of a data warehouse for ETL processes using Terraform, PostgreSQL, AWS Glue, Pyspark, S3, RDS and Airflow for a credit risk business case. | 0Data Engineer
| numpy, google cloud storage, r, pandas, postgresql, aws sns, javascript, terraform, azure, pyspark, apache hadoop, python, docker, github, gcp, matplotlib, amazon redshift, aws rds, airflow visualization, plotly, statistics, aws, aws ecr, amazon rds, mongodb, dynamodb data, aws redshift, aws glue, scikit-learn, apache airflow, flask, aws s3, dash, github actions, hadoop, aws lambda |
Developed and maintained all extractions from source data (REST API, MemSql databases, Raw files) to Snowflake DestinationOracle, using Sql and Python scripts. The proprietary legacy ELT got deprecated and the client, a real estate company, was able to use modern technologies. | 0Data Engineer
| other |
Designed, developed and maintained business intelligence solutions, dashboards and report visualization using BI tools (Amazon QuickSight, Power BI and Tableau), guaranteeing the accessibility, performance, reliability and security of the data to be consulted in an educational platform. | 0Data Engineer
| numpy, apache spark, c#, tableau, sas, pandas, powerquery, databases, dax, spark, microsoft office, python, git, powerbi, github, programming, ml frameworks, microsoft power point, sas base, snowflake, vba, seaborn, postgresql., t-sql, vba and matlab., scala, sql, visualization, talend, sap, trello, pycharm, go, visual studio code, and tableau., sql server, r, nlp, postgresql, matlab, pyspark, power bi, microsoft excel, sql seververintegrationservices, matplotlib, decision making, mysql, vs code, aws quicksight, aws, databricks, ggplot2, powerpivot, aws s3, scipy, data, jira, powerview |
Deployed and maintained machine learning, data science, and data engineering applications using Trino, Airflow, and Spark to make them usable for thousands of users through Kubernetes and AWS, and supported product and design by collaborating with cross-functional teams. | 4Software Engineer
| go, apache spark, tensorflow, kubeflow, databases, dataflow, bash, spark, postgresql, jenkins, javascript, terraform, azure, kubernetes, python, docker, git, github, dialogflow, huggingface, aws, ml frameworks, mongodb, apache airflow, airflow, data, dynamodb, github actions |
Utilized SQL, Python, PostgreSQL, and Amazon Redshift to model and create the EAQ Mart structure, enhancing data accessibility for the EAQ projects data-driven decision-making of a telecommunications company. | 0Data Engineer
| apache spark, kafka, hdfs, tableau, apache kafka, confluence, bitbucket, pandas, databases, spark, jenkins, hive, apache hadoop, python, git, powerbi, pentaho, github, gcp, aws emr, azure data factory, mongodb, apache airflow, apache impala, scala, sql, visualization, neo4j, sap, ibm, tensorflow, hbase, sql server, bash, postgresql, collaboration, sql server integration services, terraform, azure, pyspark, power bi, cloudera, decision making, mysql, linux, aws, apache hive, accountability, azure sql, databricks, aws redshift, data management, dbt, azure cosmos, impala, airflow, rpa, data, jira, hadoop, azure data lake |
Supported diverse company initiatives including evaluation of DataOps and MLOps frameworks for potential adoption. | 0Data Engineer
| feast, apache spark, tableau, pandas, jenkins, css, javascript, aws batch, python, git, github, gcp, angular, gitlab, html, aws glue, apache airflow, splunk, bigquery, sql, tablea, dynamodb, c++, mlflow, airbyte, aws lambda, google dataproc, aws ec2, aws dynamodb, gitlab ci, express.js, sql server, nlp, great expectations, postgresql, aws sns, terraform, aws eks, matlab, pyspark, docker, microsoft excel, design patterns, visual basic, mysql, aws rds, vs code, linux, statistics, aws, java, aws step functions, databricks, mssql, aws redshift, dbt, ms sql, aws s3, dataproc, data, pytest, github actions, aws kinesis |
Optimized the score calculation of an NLP-based application that helps rank the best candidates for a specific job description given their CVs using Word2Vec, and SentenceTransformer embeddings. | 3Machine Learning Engineer
| aws ec2, go, tensorflow, ml tools, bash, nlp, postgresql, docker, python, latex, mysql, golang, statistics, aws, aws redshift, scikit-learn, keras, aws s3, sql, mlflow, computer vision |
Implemented NLP models (RNN and Transformers) and text analytics for a forum questions classifier. | 2Data Scientist
| pytorch, typer, fargate, numpy, apache spark, tableau, databases, pandas, spark, javascript, cloud sql, python, git, node, powerbi, github, gcp, programming, huggingface, python libraries and, delta tables, cloud, ml frameworks, dataabricks, html, reactjs, react, apache airflow, sql, shap, dynamodb, visualization, mlflow, angularjs, sigma, aws ec2, aws fargate, tensorflow, nlp, postgresql, pyspark, neptune.ai, docker, microsoft excel, aws ecs, mysql, hugging face, ecr, fastapi, aws, ecs, aws ecr, raspberry pi, databricks, aws redshift, scikit-learn, dbt, flask, airflow, aws s3, dash, data, s3, redshift, computer vision, sckit-learn |
Developed a testing application for the assessment of candidates and employees. | 3Machine Learning Engineer
| lightgbm, apache spark, databases, spark, javascript, text classification, apache cassandra, python, differential calculus, programming, time series, ml frameworks, online tutoring, node.js, mongodb, learning, research, apache airflow, geometry, keras, linear algebra, sql, shap, dynamodb, visualization, classification, mathematica, amazon ecs, wolfram mathematica, tensorflow, microsoft word, nlp, postgresql, matlab, machine learning, docker, aws ecs, cassandra, latex, amazon redshift, fastapi, aws, learning techniques, ai, bigtable, aws redshift, control theory, scikit-learn, airflow, data, computer vision |
Worked on different projects related to customer service enhancement, RTB, churn anticipation, automatic model generation, feature importance analysis, customer segmentation, trend forecasting, Excel models migration, requestspredictions parallelization, concept analysis from unstructured text, satellite images augmentation, and image classification, as well as MLOps tasks for 3 client companies in the U.S. and Latam. | 3Machine Learning Engineer
| pytorch, go, tensorflow, spacy, tableau, databases, pandas, nlp, postgresql, javascript, sqlalchemy, pyspark, python, microsoft excel, gcp, mysql, aws, java, ml frameworks, opencv, raspberry pi, databricks, mongodb, aws redshift, scikit-learn, apache airflow, keras, airflow, rust, shap, data, visualization, mlflow, django, react |
Created alerts and monitoring schemes using Datadog and logs from Airflow. | 0Data Engineer
| aws ec2, apache spark, redshift, sql server, hive, terraform, matlab, looker, python, git, docker, governance policies, aws ecs, powerbi, github, aws iam, aws codebuild, decision making, power bi, aws rds, aws, apache hive, databricks, snowflake, aws redshift, aws glue, datadog, dbt, apache airflow, apache impala, airflow, impala, aws s3, sql, data, github actions, google sheets, sharepoint |
Developed an ansible playbook to generate historical dumps of MySQL databases massive tables (+180MM rows) to Google Cloud Storage and Bigquery on demand as a support mechanism for historical backfills at the replication pipelines for invoices data in a large fintech company. | 0Data Engineer
| numpy, apache spark, kafka, apache kafka, pandas, dataflow, dax, aws cli, kubernetes, looker, python, git, structured streaming, sac, powerbi, cloud sql, github, gcp, sql serveranalysisservices, datastream, ml frameworks, sap analytics cloud, google compute engine, snowflake, structured streaming spark, mongodb, looker studio, google vertex ai, apache airflow, sap hana, sql, visualization, spark structured streaming, django, requests, aws lambda, sap, google cloud run, firestore database, google cloud storage, google bigquery, sql server, postgresql, aws aurora, terraform, azure, pyspark, ansible, docker, microsoft excel, gcloud cli, mysql, aws rds, linux, google cloud, nielsenanswersdata, aws, apache beam, azure sql, databricks, dagster, firestore, dbt, scikit-learn, google cloud functions, airflow, aws s3, gcp iam, github actions, golangdatabases |
Applied data analysis techniques like outlier detection, missing value imputation and data distribution computation to fully analyze datasets and answer business questions. | 1Data Analyst
| numpy, oracle, pandas, beautifulsoup, postgresdata, sqlite3, css, javascript, looker, python, git, github, root, seaborn, sql, excel powerquery visualization, c++, google sheets, mathematica, bokeh, ibm, domo, domo datawarehouse, r, bash, azure, matlab, microsoft excel, matplotlib, decision making, anaconda jupyter notebook, mysql, fastapi, linux, plotly, statistics, java, sqlite, azure sql, jupyter notebook, azure cosmos, dash, scipy, beautiful soup |
Implemented a data quality in-house library using Pythons Great Expectation library, enabling validation of workflows for all data teams across the company. | 0Data Engineer
| aws ec2, luigi, apache spark, kafka, c#, sql server, apache kafka, databases, spark, prometheus, terraform, kubernetes, pyspark, ansible, apache cassandra, python, apache hadoop, cassandra, mysql, gitlab, java, aws, databricks, chef, mesos, apache airflow, redis, airflow, scala, sql, elasticsearch, data, grafana, dynamodb, visualization, jira, hadoop, delta lake |
Developed a license plate enhancement model in Python. | 3Machine Learning Engineer
| pytorch, numpy, go, tensorflow, gpu, c#, databases, pandas, bash, tpu, postgresql, javascript, cpu multithreading, azure, python, docker, git, gcp, matplotlib, mlcube, huggingface, gradio, mysql, fastapi, java, aws, ml frameworks, opencv, azure sql, keras, flask, sql, data, c++, computer vision |
Built a supervised speech parsing Wiki dataset media Commons’ XML and uploaded it for public usage, to improve speech-to-text machine learning models. After scrapping the metadata, scripts were made to download and process audios and transcriptions using both ffmpeg and a GCP instance, which were later uploaded to GCP Cloud Storage and transferred to a HuggingFace repository using Git LFS for an open engineering consortium. | 3Machine Learning Engineer
| aws ec2, yolov4, numpy, tensorflow, google cloud storage, xgboost, pandas, nlp, postgresql, labview, python, docker, git, aws ecs, gcp storage, gcp, huggingface, tensorflow 2.0, fastapi, catboost, statistics, aws, ml frameworks, aws ecr, visual, node.js, nodejs, scikit-learn, aws sqs, apache airflow, keras, flask, aws s3, sql, data |
Developed entity-relationship diagrams (ERDs) and data flow diagrams (DFDs) to document data requirements and relationships. | 0Data Engineer
| other |
Prepared data and performed feature engineering using Vaex, MongoDB, PyMongo, Pyodbc, SQLServer, Psycopg2, and AWS Redshift for pricing and client churn probability measurements, enabling customers insights and informed decisions making on client retention. | 2Data Scientist
| numpy, kafka, apache kafka, redux, databases, pandas, perfect, css, javascript, heroku, python, github, gcp, selenium, programming, time series, ml frameworks, html, node.js, software architecture, mongodb, apache airflow, keras, spring, planning, pymongo, ruby on rails, react.js, visualization, postgres, travisci, django, teamwork, bokeh, graphql, go, express.js, tensorflow, xgboost, r, ownership, storybook, streamlit, bash, postgresql, collaboration, pyspark, docker, microsoft excel, sqlserver, mysql, plotly, statistics, aws, nginx, java, c, opencv, jquery, typescript, aws redshift, scikit-learn, sidekiq, flask, airflow, data, pytest, sinatra, ruby, php, react |
Built machine learning models for customer segmentation. | 1Data Analyst
| other |
Created a Python routing package using OR-Tools and OSRM that improved collaboration among developers. | 2Data Scientist
| numpy, apache spark, pandas, aws step function, microsoft sql server, python, git, powerbi, github, data modeling, aws glue, apache airflow, sql, dynamodb, google sheets, airbyte, aws lambda, aws ec2, sql server, bash, postgresql, aws sns, collaboration, terraform, azure, pyspark, docker, power bi, aws rds, linux, statistics, aws, airtable, azure sql, databricks, mssql, dagster, aws redshift, scikit-learn, dbt, shell scripts, aws s3, aws dynamodbdata, github actions, aws kinesis, scikit learn |
Analyzed more than 500 tables and ETLs using DataCleaner to create data quality standards across the whole organization. | 1Data Analyst
| powercenter, ibm, google bigquery, oracle, sql server, r, sas, databases, google analytics, sql server integration services, python, powerbi, datacleaner, gcp, fivetran, decision making, asana, aws glue visualization, aws, google tag manager, oracle database, sql ssis, snowflake, aws redshift, aws glue, asp.net, dbt, apache airflow, sas miner, google cloud functions, airflow, ibm datastage, aws s3, sql, google data studio, data, aws lambda |
Developed and maintained a dailyOps monitoring tool using Python, SQL, Git, Pandas, StreamLit, NumPy, Plotly and Ubuntu to detect and mitigate problems related models on production for an industrial start-up. | 2Data Scientist
| hyperopt, numpy, lightgbm, apache spark, spacy, sas, pandas, databases, spark, python, git, powerbi, pycaret, gcp, nltk, time series, ml frameworks, vba, seaborn, keras, sql, dynamodb, visualization, mlflow, aws ec2, tensorflow, xgboost, google bigquery, r, streamlit, nlp, facebook prophet, prophet, azure, pyspark, microsoft excel, matplotlib, decision making, azure devops, linux, plotly, aws, statsmodels, databricks, evidently, aws redshift, sagemaker, scikit-learn, scipy, google data studio, data |
My team handles over 1.5 Billion registries per day | 4Software Engineer
| apache activemq, aws cloudformation, oracle, activemq, rabbitmq, jenkins, javascript, subversion, git, gcp, aws emr, aws athena, virtuoso, resque, chef, sql, dynamodb, django, iaas, kibana, graphql, wireshark, jmeter, postgresql, docker, docke, scalatest, linux, unix, paas, aws redshift, lisp, pytest, hadoop, java dropwizard, php, junit, apache spark, databases, pandas, spark, logstash, heroku, rails, apache hadoop, python, github, ml frameworks, apache ab, mongodb, microsoft project, erlang, scala, elasticsearch, visualization, celery, aws lambda, aws ec2, go, bash, scrapy, perl, ansible, mysql, aws rds, vs code, statistics, java, sqlite, aws, c, intellij, puppet, flask, redis, aws s3, data, jira, clojure, python scrapy, react |
Implemented an improved CICD and automated testing system leveraging Snowflake cloning, AWS, Matillion and various APIs reducing average time to production and increasing developer confidence for data services team of a furniture retailer. | 0Data Engineer
| pytorch, spacy, circleci, sodavisualization, rabbitmq, jenkins, javascript, ms sqlserverdata, python, git, powerbi, gcp, angular, ml frameworks, matillion, snowflake, mongodb, soda, apache airflow, keras, sql, shell, c++, django, tensorflowextended, raspberrypi, tensorflow, google bigquery, sql server, r, postgresql, terraform, docker, power bi, vue, amperity, swagger, linux, fastapi, java, aws, c, airtable, typescript, octave, kotlin, firebase, shell scripts, redash, yaml, flask, airflow, rust, grafana, mercurial, arduino, express, computer vision, react |
Writing applications with a good design to be delivered continuously and follow deployment status to other teams (security, performance, etc) and then final users. | 4Software Engineer
| decrease, data in a very intense batch job., apache spark, kafka, oracle, rest api, apache kafka, kubeflow, i was asked to design and build tools to improve citizens relationship, databases, with government, performance, spark, jenkins, javascript, bureaucracy and other key aspects to improve government efficiency., kubernetes, microsoft office, writing applications with a good design to be delivered continuously and follow deployment status to other teams, python, software engineer, using java related technologies, senior software engineer, my tasks were related to deliver solutions that would help, my main tasks were, gitlab, ml frameworks, programs to decrease debts, software architecture, people across the state., snowflake, applying multithreading techniques, pulsar, apache airflow, applying best practices and selecting tools that help developing and delivering applications in a fast-paced and reliable way.
programming, softtek jan 2012, scala, acting mainly as an integration architect inside one of the great retailers on brazil, resource it solutions jan 2013, visualization, mlflow, metabase, sigma, go, working for a state sponsored company, magna sistemas sep 2013, postgresql, the team were able to almost double the throughput of processed, azure, delivered solutions to support a large number of accesses by users worldwide, using an agile methodology and java for a large financial company., process and transform external data delivered by third-party providers into useful, mysql, golang, java, aws, data inside the customer. the main project was maintain a home made solution to integrate hr data, that later was replaced by a solution designed on oracle odi. ive participated designing the solution, applied tdd and architecture knowledge to increase code coverage, testing and fixing errors on odi scripts., deliver information, dbt, increase transparency, tech mahindra may 2014, redash, with a high availability and secure architecture is my current focus, airflow, aws s3, data, s3, related to receive, seldon |
Leveraged visualizations in R and ggplot2 as a tool for communicating results, introducing SHAP Values to the Analytics team to extract feature importance and explain black-box models, helping the bank use robust complex models and understand their outputs. | 2Data Scientist
| caret, lightgbm, apache spark, r, databases, swift, nlp, spark, raster, pyspark, microsoft office, looker, python, git, microsoft excel, shiny, matplotlib, decision making, time series, plotly, statistics, ml frameworks, microsoft power point, databricks, vscode, jupyter notebook, seaborn, ggplot2, scikit-learn, bigquery, sql, shap, data, visualization, microsoft powerpoint |
Created profitability and optimization models. | 2Data Scientist
| pytorch, apache spark, jupyter notebooks, c#, tableau, pandas, javascript, apache hadoop, python, apache cassandra, git, gcp, time series, snowflake, mongodb, tableau and power, apache airflow, nosql, sql, snowflake.data, neo4j, tensorflow, fiddler, r, azureml frameworks, bash, bentoml, nlp, postgresql, powershell, azure, pyspark, matlab, docker, aws ecs, cassandra, visual basic, mysql, fastapi, aws, aws ecr, yatai, numpyvisualization, azure sql, databricks, jupyter notebook, scikit-learn, dbt, shell scripts, matlab and visual basic.databases, computer vision |
Created dynamic dashboards and reports with real-time and forecast analytics using Python, SciPy, What-if, and Oracle Big Data Appliance. | 0Data Engineer
| oracle, circleci, ibm db2, looker, hyperion oracle, powerbi, gcp, data modeling, terremark dc, aws athena, aws emr, ec2, aws glue, sql, ibm bluemix, sap, ibm, tensorflow, google cloud storage, postgresql, terraform, sqlalchemy, azure, docker, power bi, google cloud, paas, mssql, jupyter notebook, aws redshift, shell scripts, scipy, hadoop, aws kinesis, apache kafka, apache hadoop, python, apache cassandra, cloud sql, github, sql serveranalysisservices, time series, ibm softlayer, html, ms sql server, ebs, progress, mongodb, apache airflow, scala, ms access, aws ebs, .net, aws lambda, aws ec2, db2, hbase, sql server, r, oracle cloud, pyspark, ansible, windows azure, mysql, aws rds, statistics, aws, java, c, spotfire, apache hive, aws step functions, azure sql, t-sql microsoft, windows, puppet, aws s3 |
Created a streamlined end-to-end pipeline, refining broken chunks of code, to improve an existing load recommender XGBoost model. Managed to boost test AUC from 0.83 to 0.93 (~custom metric boost of 62%). Deployed web service endpoints. | 3Machine Learning Engineer
| pytorch, numpy, databases, pandas, jenkins, kubernetes, python, git, gcp, data modeling, snowflake, seaborn, google big query, apache airflow, keras, sql, visualization, mlflow, tensorflow, xgboost, google bigquery, bash, nlp, facebook prophet, terraform, azure, pyspark, docker, ml flow, matplotlib, fastapi, aws ml, java, aws, azure sql, scikit-learn, flask, airflow, data, jira, computer vision |
Developed action plans to prevent the deterioration of the investment made by the bank. | 1Data Analyst
| numpy, apache spark, c#, tableau, sas, pandas, powerquery, databases, dax, spark, microsoft office, python, git, powerbi, github, programming, ml frameworks, microsoft power point, sas base, snowflake, vba, seaborn, postgresql., t-sql, vba and matlab., scala, sql, visualization, talend, sap, trello, pycharm, go, visual studio code, and tableau., sql server, r, nlp, postgresql, matlab, pyspark, power bi, microsoft excel, sql seververintegrationservices, matplotlib, decision making, mysql, vs code, aws quicksight, aws, databricks, ggplot2, powerpivot, aws s3, scipy, data, jira, powerview |
Designing, testing and maintaining data management in Snowflake. | 0Data Engineer
| other |
Developed ETL pipelines with SQL to feed dashboards for BI insight extraction. | 1Data Analyst
| numpy, google bigquery, tableau, oracle, r, pandas, kimball-style data modeling, sqlite data, databricks visualization, looker, python, microsoft excel, gcp, alation, data modeling, decision making, time series, statistics, sqlite, spss, databricks, snowflake, vba, stata, java databases, sql, medallion architecture |
Designed and implemented a Lakehouse architecture from scratch on AWS, enabling data consumption and accurate insights for the business, helping to pivot of the product. The architecture and underlying infrastructure decisions had a low operational cost but a ready-to-scale approach. Stack: AWS S3, Lambda, EMR on EKS, Glue, Athena, Airflow, Redshift. Apache Spark. Scala, Python. Gitlab CICD. | 0Data Engineer
| fluentd, fargate, redshift., apache spark, oracle, built power bi dashboards are helping with the communication of insights and monitoring of kpis using dbt., databases, pandas, data engineer, spark, lambda, starburst, hive, kubernetes, python, git, powerbi, support and extension of the company data lake, github, emr on ec2, apache spark. scala, aws emr, aws athena, gitlab, ml frameworks, from which the app team could monitor and discover patterns without increasing costs significantly., designed the data model for the data warehouse with a strong focus on product features and its usage., kinesis data streams, mongodb, oracle rdbms, python., aws glue, apache airflow, and being assigned as the data engineer for several strategic projects b2b and b2c of the company., scala, sql, dynamodb, visualization, aws lambda, aws ec2, atlan, go, sql server, belcorp jan 2018, aws eks, sql., aws ecs, kyligence, with serverless querying, including an in-house coordinator app for custom business rules that were not supported by orchestration tools in that moment., kyvosdeveloped a real-time pipeline for app transactions, linux, java, aws, delivering computed metrics and user insights into several in-house destinations such as sales applications and machine learning models., apache hive, part of the big data team within the data, athena. python, kinesis data firehose, data analysis with sql to test product hypothesis results and their impact on growth., aws redshift, dbt, airflow, aws s3, development of etls, data, s3, aws kinesis, redshift |
Developed and deployed unsupervised learning models for customers segmentation based on their financial behavior. Implemented ML Models state of art. | 2Data Scientist
| other |
Used optimization techniques such as integer programming to solve the Meal Delivery Routing problems. | 2Data Scientist
| pytorch, apache spark, cloudwatch, rest api, r, jenkins, prometheus, mongo db, d3, aws cloudwatch, python, git, github, decision making, time series, fastapi, statistics, aws, ml frameworks, databricks, snowflake, mongodb, sklearn, d3.js, datadog, sparkvisualization, sql, data, c++, github actions |
Implemented a real-time streaming application to perform transformations and warehousing of insurance payments by processing over a million transactions per week for one of the largest finance corporations in the Americas, using Kafka, Azure Event Hub, ADF, and Azure SQL DB. | 0Data Engineer
| apache spark, kafka, rest api, apache kafka, databases, pandas, spark, python, github, aws athena, ml frameworks, jupyter, aws glue, seaborn, apache airflow, scala, sql, visualization, django, transact-sql, azure synapse analytics, streamlit, azure, azure synapse, aws ecs, matplotlib, aws, azure sql, databricks, jupyter notebook, aws redshift, dbt, aws sqs, zabbix rest api, airflow, data, github actions, azure databricks, react |
Led the development of a soccer match outcome forecasting method for sports analytics using ConvNets. | 2Data Scientist
| pytorch, apache spark, spark, apache cassandra, python, github, data modeling, aws emr, ml frameworks, mongodb, aws glue, seaborn, apache airflow, planning, sql, dynamodb, visualization, aws lambda, aws ec2, tensorflow, r, geospatial analysis, bash, nlp, postgresql, aws sns, terraform, sqlalchemy, matlab, retool, pyspark, docker, matplotlib, decision making, mysql, aws rds, fastapi, statistics, aws, aws step functions, h2o automl, aws redshift, ggplot2, scikit-learn, dbt, airflow, aws s3, dash, data, gis, github actions, aws kinesis, computer vision |
Designed and developed the frontend and the backend for a health mobile app to book medical appointments using Django and Django REST in the back end and HTML5, CSS, Bootstrap and jQuery in the frontend. | 4Software Engineer
| other |
Designed warehouse layouts based on information extracted from orders, dispatches, budget, and available physical and human resources. In addition, choosing the best packing and picking methodology to suit the customer involved, reducing the productivity and costs of logistics operations significantly. | 1Data Analyst
| pytorch, typer, fargate, numpy, apache spark, tableau, databases, pandas, spark, javascript, cloud sql, python, git, node, powerbi, github, gcp, programming, huggingface, python libraries and, delta tables, cloud, ml frameworks, dataabricks, html, reactjs, react, apache airflow, sql, shap, dynamodb, visualization, mlflow, angularjs, sigma, aws ec2, aws fargate, tensorflow, nlp, postgresql, pyspark, neptune.ai, docker, microsoft excel, aws ecs, mysql, hugging face, ecr, fastapi, aws, ecs, aws ecr, raspberry pi, databricks, aws redshift, scikit-learn, dbt, flask, airflow, aws s3, dash, data, s3, redshift, computer vision, sckit-learn |
Implemented and maintained ELT pipelines with Python to extract data in batch (daily) from sources as Firestore (e-commerce), SQL Server (sales), CSV and Excel (marketing data) to feed Azure data lake. | 0Data Engineer
| numpy, apache spark, kafka, apache kafka, pandas, dataflow, dax, aws cli, kubernetes, looker, python, git, structured streaming, sac, powerbi, cloud sql, github, gcp, sql serveranalysisservices, datastream, ml frameworks, sap analytics cloud, google compute engine, snowflake, structured streaming spark, mongodb, looker studio, google vertex ai, apache airflow, sap hana, sql, visualization, spark structured streaming, django, requests, aws lambda, sap, google cloud run, firestore database, google cloud storage, google bigquery, sql server, postgresql, aws aurora, terraform, azure, pyspark, ansible, docker, microsoft excel, gcloud cli, mysql, aws rds, linux, google cloud, nielsenanswersdata, aws, apache beam, azure sql, databricks, dagster, firestore, dbt, scikit-learn, google cloud functions, airflow, aws s3, gcp iam, github actions, golangdatabases |
Delivered technical consulting. | 4Software Engineer
| pytorch, numpy, go, tensorflow, gpu, c#, databases, pandas, bash, tpu, postgresql, javascript, cpu multithreading, azure, python, docker, git, gcp, matplotlib, mlcube, huggingface, gradio, mysql, fastapi, java, aws, ml frameworks, opencv, azure sql, keras, flask, sql, data, c++, computer vision |
Developed a customer churn predictive model using SQL, R, STATA and Tableau. This prediction allowed the company to build different retention strategies to avoid good customers leaving the company. | 1Data Analyst
| pytorch, gcp ml tools, lightgbm, tensorflow, google cloud storage, xgboost, tableau, oracle, r, streamlit, sas, nlp, postgresql, apache hadoop, python, git, docker, gcp, decision making, amazon redshift, mysql, fastapi, statistics, aws, html, node.js, aws redshift, vertexai, stata, scikit-learn, keras, apache impala, bigquery, sql, shap, data, hadoop, pytorch lightning |
Developed an AutoML script using Python to generate challenger models to test the quality of the different models being validated for one of the biggest insurance companies in the US. | 2Data Scientist
| aws ec2, numpy, tensorflow, google bigquery, r, streamlit, databases, pandas, bash, compute engine, pool, postgresql, cloud sql, python, git, docker, microsoft excel, aws ecs, gcp, matplotlib, decision making, aw ec2, mysql, fastapi, cloud, statistics, aws, spss, google compute engine, scikit-learn, google vertex ai, excel, aws mltools, asyncio, google cloud functions, bigquery, sql, shap, data, lime |
Developed and deployed cloud-based RPA solutions and machine learning models using Python and REST APIs for clients in the finance, retail and education sector. | 4Software Engineer
| apache spark, beautifulsoup, css, javascript, kubernetes, python, powerbi, github, gcp, data modeling, html, delta lake, aws glue, seaborn, apache airflow, splunk, sql, dynamodb, django, aws lambda, aws ec2, aws dynamodb, tensorflow, google cloud storage, google bigquery, postgresql, aws sns, terraform, pyspark, docker, microsoft excel, matplotlib, mysql, aws rds, linux, aws, aws step functions, apache hive, typescript, databricks, jupyter notebook, aws redshift, dbt, flask, aws s3, dash, rpa, data, beautiful soup, github actions, aws kinesis, delta live tables, react |
Integrated a secure payment gateway into the vacation clubs system, using Java Spring on the backend to process transactions of substantial value on a regular basis. | 4Software Engineer
| numpy, kafka, apache kafka, redux, databases, pandas, perfect, css, javascript, heroku, python, github, gcp, selenium, programming, time series, ml frameworks, html, node.js, software architecture, mongodb, apache airflow, keras, spring, planning, pymongo, ruby on rails, react.js, visualization, postgres, travisci, django, teamwork, bokeh, graphql, go, express.js, tensorflow, xgboost, r, ownership, storybook, streamlit, bash, postgresql, collaboration, pyspark, docker, microsoft excel, sqlserver, mysql, plotly, statistics, aws, nginx, java, c, opencv, jquery, typescript, aws redshift, scikit-learn, sidekiq, flask, airflow, data, pytest, sinatra, ruby, php, react |
Built an in-house end-to-end solution chatbot platform using Python to design, create, and deploy bots for one of the most important financial institutions in Colombia and its Top 2partner banking startup in Panama. The solution was deployed using AWS EC2 machines and replaced the use of standard cloud-based platforms for handling conversations, reducing the time and money expended. | 2Data Scientist
| pytorch, networkx, numpy, c#, pandas, python, git, gurobi optimization solver, scikit-image, ml frameworks, html, distributed systems, seaborn, keras, apache impala, sql, dynamodb, c++, aws ec2, aws dynamodb, tensorflow, bash, cloudera data lake, matlab, pyspark, matlab-simulink, impala sql, matplotlib, ros, latex, java, aws, c, opencv, scikit-learn, shell scripts, tf object detection api, flask, aws s3, data, open cv, computer vision |
Constructed a data processing pipeline using Airflow, Python, and NodeJS to prepare and load videos of exercise routines into a data lake on AWS S3. The resulting data was used to evaluate ML models and perform exploratory data analysis. | 2Data Scientist
| pytorch, docker compose, numpy, tableau, rest api, databases, pandas, css, javascript, python, librosa, git, compose, github, tfjs, ml frameworks, html, tfx, node.js, mongodb, tf serving, dvc, tensorflowjs, nodejs, apache airflow, keras, tensorflow serving, sql, c++, mlflow, postgres, aws lambda, aws ec2, tensorflow, r, ownership, nlp, postgresql, scrapy, collaboration, terraform, sqlalchemy, docker, aws ecs, aws vpc, mysql, fastapi, plotly, statistics, aws, numpyvisualization, opencv, autogluon, kotlin, scikit-learn, airflow, aws s3, dash, data, pytest, github actions, computer vision, react |
Collaborated with 3 business departments (finance, sales, and risk) to align the best way to achieve the proposed goals, advising regarding ML models development | 2Data Scientist
| pytorch, eks., numpy, tox, aws elastic beanstalk, c#, tableau, databases, pandas, metaflow, javascript, diffusers, kubernetes, python, git, github, selenium, huggingface, time series, aws - rds, ml frameworks, ec2, snowflake, seaborn, apache airflow, keras, sql, visualization, mlflow, postgres, travisci, beatifulsoap, aws ec2, tensorflow, xgboost, streamlit, postgresql, php, iam, aws eks, docker, aws ecs, aws iam, matplotlib, mysql, hugging face, aws rds, ecr, fastapi, detectron airflow, statistics, aws, ecs, aws ecr, opencv, aws redshift, scikit-learn, dbt, aws s3, scipy, data, pytest, github actions, computer vision |
Created a one-stage object detector using the YOLOV4 model and served it through a dockerized API deployed in AWS. | 3Machine Learning Engineer
| gcp cloud functions, apache spark, databases, spark, pruning, java programming, anomaly detection, python, powerbi, gcp, programming, and nodejs, data analytics, cloud infrastructure, nltk, cloud, neural network, ml frameworks, node.js, programming language, aws glue, glue, big data processing, keras, artificial intelligence, sql, rds and, visualization, c++, aws ec2, gcp cloud storage, unsupervised learning, go, tensorflow, google cloud storage, amazon s3, gcp compute engine, text analysis, nlp, postgresql, gcp app engine, scrapy, image processing, terraform, machine learning, power bi, java and, other, mysql, aws rds, tensorflow 2.0, fastapi, java, aws, ai, and dash., data structures, scikit-learn, api, and athena., data cleaning, flask, aws s3, data, and keras. |
Developed VBA applications and RPA processes to automate emails sending, data extraction and transformation, reports updating reports, among others, that eliminated multiple resources involvement and improved performance and productivity. | 1Data Analyst
| microsoft visio, sap, qlikview, ibm, r, rpa tools, statgraphics, powerquery, visio, qlik, dax, power pivot, python, minitab, microsoft excel, powerbi, decision making, data modeling, minitab power bi, alteryx, c, power query, data studio, snowflake, vba, excel, planning, powerpivot, sql, rpa, data, prepbuilder, visualization, sap erp, dax syntax |
Managed projects related to migration from On-Prem (Teradata) to the Cloud (AWS, GCP). | 0Data Engineer
| other |
Implemented statistical methods to make data-driven technical decisions. | 0Data Engineer
| apache spark, aws cloudformation, oracle, sql servers, dax, spark, dax., cloud sql, python, git, powerbi, programming, rds, aws emr, cloud, gitlab, matillion, azure data factory, dbschema, apache airflow, big data processing, sql, visualization, aws lambda, sap, azure analysis services., bash, postgresql, azure, docker, azure data warehouse., aws rds, aws, lambda. azure data factory, gradle., azure sql, kotlin, aws redshift, aws s3, kotlin., power bi., redshift, emr, oracle dbs |
Involved in the analysis and definition of a methodology that statistically determines the success of trendy colors in lingerie. | 1Data Analyst
| other |
Created APIs for both models using FastAPI and deployed Docker containers on AWS Lambda. | 3Machine Learning Engineer
| pytorch, numpy, lightgbm, rest api, sas, pandas, databases, aws batch, python, github, huggingface, nltk, time series, ml frameworks, aws sdk for python (boto3), dvc, sql, aws eventbridge, aws api gateway, c++, mlflow, aws lambda, aws ec2, go, ibm, tensorflow, xgboost, streamlit, nlp, aws sns, terraform, matlab, pyspark, docker, aws ecs, matplotlib, fastapi, statistics, aws, nginx, aws ecr, aws redshift, scikit-learn, aws s3, aws sdk for python, data, github actions, computer vision |
Developed Customer lifetime value code to understand campaigns performance and deliver data-based decisions related to marketing strategies using SQL, Excel and Python. | 1Data Analyst
| numpy, jupyter notebooks, oracle, hdfs, pandas, apache hadoop, python, cloud sql, powerbi, r studio, gcp, nltk, ml frameworks, snowflake, mongodb, sql, visualization, periscope, tensorflow, r, j google colab, nlp, oracle cloud, postgresql, power bi, microsoft excel, mixpanel, matplotlib, mysql, rstudio, linux, statistics, apache hive, databricks, vscode, jupyter notebook, apache nifi, ggplot2, scikit-learn, google data studio, data, hadoop |
Developed operational dashboards using Tableau leveraging on visualizations and scheduled alerts to remedy small urgent needs in daily operations. | 1Data Analyst
| spacy, python and power query helping to reduce processing times and providing quality in data visualizations., powerquery, and integrated a data model for the companys general budget using google sheets leveraging cloud functions and dataset connections., microsoft office, looker, powerbi, supported the implementation of the erp of the company with data transformation and data migration using excel and vba macros achieving to pass from 5 different systems to an unique enterprise system., neural networks, algorithms, azure data factory, snowflake, financial analyst, tableau desktop, nosql, urbanas, sql, provided indicators to the sales force through specialized reports using looker studio and sac for daily operations monitoring., transformed data and created models for analysis with tools such as sql, foundations, sap, finance, finance., and strategy for decision-making of a micro-financial organization achieving to create important products such as bonus calculation tool, advanced, postgresql, azure, machine learning, power bi, microsoft excel, decision making, databricks data analyst associate certification, and presented dashboards to the executive team., performed financial analysis and supported business cases for initiatives and new projects., led the business intelligence team and created, databricks, created and managed sales and financial forecasting tools., to generate findings in satisfaction surveys for the customer journey team., hr, supported, created specialized tools in excel and vba macros to support daily operations in different company departments such as cost, among other things., specialization, tableau, developed an etl workflow using python, managed, pandas, business administration, customer journey analytical tool, process analyst, cloud sql, fundación de la mujer, python, sac, applied machine, sap analytics cloud, leveraging api data extraction and text processing libraries, looker studio, vba, pandasvisualization, artificial intelligence, implemented reports for strategic analysis and monitoring kpis using business objects and looker studio., bi lead, google sheets, built operational dashboards using power bi leveraging on report tables and trend charts to provide the status and details of all the orders to the purchase department, santander university, intermediate level, python databases, sql server, nlp, created, education, marketing, azure sql, human resources, budget tracking and forecasting tools for each branch office and management team, dbt, data, implemented an etl pipeline using azure data factory, created interactive dashboard for a project using databricks leveraging on data transformation to gain some insights into the purchasing behavior of a households dataset, basic aspects, leveraging azure storage to create report tables for analysis from raw data. |
Designed and developed a metadata framework using Databricks Delta Live Tables to allow users of a Fortune 500 consulting company to create data pipelines following a configuration-driven approach. The framework supported streaming and batch processing, ingesting data from multiple data sources (including Kafka, text files, JDBC connections), real-time data quality expectations, SQL-based transformations, and a plugin environment to add custom code, allowing to process over 1000 tables in 1 month. This work led to an invitation by Databricks to become a Databricks Solutions Architect Champion. | 0Data Engineer
| apache spark, kafka, apache kafka, javascript, python, git, structured streaming, powerbi, gcp, nltk, ml frameworks, snowflake, delta lake, seaborn, apache airflow, sql, visualization, redshiftdata, aws dms, spark structured streaming, aws lambda, r, nlp, great expectations, postgresql, sqlalchemy, docker, matplotlib, mysql, snowsight, linux, aws, statsmodels, raspberry pi, databricks, scikit-learn, databricks jobs api, flask, gensim, airflow, aws s3, delta live tables |
Implemented backend and front-end frameworks to display credit default risk predictions and explainability using Streamlit and LIMESHAP libraries. | 3Machine Learning Engineer
| pytorch, numpy, circleci, aws batch, kubernetes, microsoft office, huggingface transformers, git, huggingface, aws emr, neptune, node.js, snowflake, seaborn, pinecone, keras, langchain, prefect, sql, shap, dynamodb, prioritization, mathematica, javadatabases, aws dynamodb, tensorflow, dask, hugingface peft, postgresql, terraform, aws eks, docker, aws ecs, matplotlib, decision making, latex, fastapi, linux, aws quicksight, unix, aws redshift, scikit-learn, aws sqs, tf object detection api, github actions, hadoop, aws neptune, aws elastic beanstalk, tableau, aws s3data, pandas, oidc sso connectors, apache cassandra, python, apache hadoop, github, chainlit, sql serveranalysisservices, open ai gpt4, time series, ml frameworks, mongodb, daskvisualization, nodejs, apache airflow, aws api gateway, mlflow, aws ec2, streamlit, bash, nlp, scrapy, pyspark, mysql, aws rds, aws, java, c, aws ecr, aws ssm, dbt, flask, aws s3, dash, traefik |
Analyze, design and implement a monitor for the material discharge hoppers in the packing area using BPM Metastorm, .NET and SQL Server. | 1Data Analyst
| numpy, oracle, universidad de san buenaventura, mongodb developer and administrator, hive, microsoft certified technology specialist .net framework 3.5 asp.net applications, synapse, part of the pmo team using asp.net, git, apache storm, eclipse debugging techniques and tricks, universidad internacional de valencia, data modeling, aws emr, angular, implementation and maintenance, jupyter, yarn, scrum master certified expert, azure data factory, prestosql, snowflake, big data and data science, microsoft certified technology specialist sql server 2008, aws glue, apache impala, nosql, analyzed, training and live output for the sap casa 2008 project, sql, itil foundations v3, sap, apache spark 2.0 with scala on with big data, courses, sql server integration services, bigml, terraform, python for data analysis and visualization, azure, azure synapse, aws cloudwatch, integration test, pmp, microsoft certified professional, databricks, jupyter notebook, complete python bootcamp, aws redshift, scikit-learn, metastorm business process management v9 delta, sklearn for predictive data analysis, itil, using bigml, hadoop, and statsmodels to develop statistical models.bsc, azure data lake, apache spark, hdfs, tableau, apache kafka, sqoop, pandas, scrum master, big data hadoop architect master, qlik, spark, azure blob storage, apache hadoop, python, apache cassandra, ektron cms developer, drugs and weapons in the city of madrid, ssrs, gitlab, hue, ml frameworks, software architecture, mongodb, generated indicators in the execution phases, javascriptdatabases, apache airflow, big-data and hadoop developer, scala, visualization, .net, impala-sql engine for hadoop, apache ranger, angular 4, go, sql server, zeppelin, developed and implemented a hazardous materials management solution used in the company using asp and sql server, eclipse, pyspark, bpm, systems engineering, analyzed influential factors in crimes related to consumption of alcohol, mysql, education, aws, java, ranger, apache hive, asp.net, airflow, aws s3, presto, data, designed, aws big data speciality |
Built a Tableau dashboard to enable presentations to stakeholders of a bike share system, giving them access to real-time data. | 1Data Analyst
| qlikview, teradata, apache spark, oracle, tableau, databases, pandas, powerquery, dax, apache hadoop, python, powerbi, alation, miro, snowflake, microsoft project, seaborn, databricksvisualization, sql, microsoft visio, trello, ibm, tensorflow, domo, netezza, r, bash, collaboration, power bi, microsoft excel, matplotlib, asana, linux, plotly, c, databricks, ggplot2, shell scripts, redash, dash, google data studio, data, jira, leadership |
Conducted four Proof of Concept to assess new software vendors for the data ecosystem. Worked with international engineers evaluating performance, suitability and user experience of the products. Stack: Kyligence, Kyvos, Starbust, Atlan. | 0Data Engineer
| fluentd, fargate, redshift., apache spark, oracle, built power bi dashboards are helping with the communication of insights and monitoring of kpis using dbt., databases, pandas, data engineer, spark, lambda, starburst, hive, kubernetes, python, git, powerbi, support and extension of the company data lake, github, emr on ec2, apache spark. scala, aws emr, aws athena, gitlab, ml frameworks, from which the app team could monitor and discover patterns without increasing costs significantly., designed the data model for the data warehouse with a strong focus on product features and its usage., kinesis data streams, mongodb, oracle rdbms, python., aws glue, apache airflow, and being assigned as the data engineer for several strategic projects b2b and b2c of the company., scala, sql, dynamodb, visualization, aws lambda, aws ec2, atlan, go, sql server, belcorp jan 2018, aws eks, sql., aws ecs, kyligence, with serverless querying, including an in-house coordinator app for custom business rules that were not supported by orchestration tools in that moment., kyvosdeveloped a real-time pipeline for app transactions, linux, java, aws, delivering computed metrics and user insights into several in-house destinations such as sales applications and machine learning models., apache hive, part of the big data team within the data, athena. python, kinesis data firehose, data analysis with sql to test product hypothesis results and their impact on growth., aws redshift, dbt, airflow, aws s3, development of etls, data, s3, aws kinesis, redshift |
Developed an automated tool for documenting data assets by leveraging Atlans REST API to import data asset descriptions, using git for versioning, and hosting a dbt project. This improved the efficiency of the documenting process for the data teams of one of the biggest fintech companies in Mexico. | 0Data Engineer
| apache spark, rest api, spark, apache cassandra, python, git, github, selenium, matillion, snowflake, mongodb, aws glue, apache airflow, pymongo, apache flink, bigquery, scala, sql, redshiftdata, metabase, aws lambda, bash, great expectations, pyspark, docker, microsoft excel, design patterns, aws rds, oop, aws, aws step functions, databricks, aws redshift, dbt, airflow, aws s3, data, github actions, aws kinesis, solid principles |
Designed and implemented bonus payment models for the commercial and debt collection employees to align their motivation with the business goals. | 1Data Analyst
| sap, power automate, tableau, microsoft word, pandas, jinja, python, git, sharepoint, microsoft excel, powerbi, github, decision making, sap business objects data services, sap business objects web, vs code, microsoft power point, ms excel, minitab visualization, snowflake, mssql, oracle data, dbt, apache airflow, ms sql, planning, sql, sap business objects bi, matlab databases, power apps, sigma |
Built a text classification model based on the DistilBERT transformer architecture using HuggingFace, Tensorflow, and MLflow. The model aimed to classify newly created social media posts as either real or fake to prevent the harmful effects of their rapid propagation. The chosen transformer architecture enabled the model to make quick and accurate predictions in real-time, which was essential in an environment where vast amounts of data were constantly being produced. | 3Machine Learning Engineer
| removal of punctuation, and a vector indexing search module based on the approximate nearest neighbor algorithm that allows for accurate and faster searching of the k closest embedding vectors given a query embedding vector in large data sets. developed a backend application that utilized fastapi, pyqt, numpy, spacy, n. valencia-marulanda, projects, convolutional neural networks, matlotlib, o. aguirre-ospina, and tensorflow to provide targeted interventions and support to students at risk of dropping out of the university, a platform over which the desktop application ran, developed characterization schemas based on the u-net computer vision model and statistical measurements using keras, huggingface, implemented a rest api using fastapi to serve the text classification model mentioned above.created an etl pipeline using numpy, clean, boto3, national university of colombia feb 2021, 17th ibero-american conference on ai, 2023., proceedings., m. calderón-marulanda, correlated predictors, dvc, seaborn, splitting of contractions, enabling medical teams to objectively estimate epidural anesthesia effectiveness and diagnose patients quickly and accurately., keras, and optimization, a. alvarez-meza, sql, a low-cost, including normalization, mateo tobón-henao, sales, and stopwords.built a text classification model based on the distilbert transformer architecture using huggingface, tensorflow, and aws api gateway to serve the aforementioned loan default prediction model., postgresql, national university of colombia july 2021, national university of colombia, instituto politécnico nacional, matlab, which was essential in an environment where vast amounts of data were constantly being produced., mne, machine learning, docker, graduated with honors., matplotlib, openai api, using python. designed a desktop application designed using matplotlib, fastapi, for thermal images capturing and analyzing. included raspad, spellchecker, andrés álvarez-meza, d. collazos-huertas, developed a student dropout classification model using numpy, designed representation methodologies through functional connectivities using scipy, and aws lambda, structuring machine learning projects, data engineering nanodegree program, a python-based computer vision tool to support anesthesia assessment procedures using infrared thermography. advances in artificial, enabling users to interact with the aforementioned search engine.created an etl pipeline using nltk, ba, scikit-learn, the complete python course, november, numbers, improve their retention rates, and tensorflowlite, scipy, and efficiency in restaurants. the search engine including a preprocessing module based on the siamese vision transformer and openais clip models that project image-to-image and text-to-image databases into a low-dimensional latent space where the similarity between paired samples in the original space is preserved., hyperparameter tuning, developed preprocessing methods based on spatial filters such as surface laplacian and independent component analysis to reduce the low signal-to-noise ratio, 2022, spotifys annoy, rest api, and the encoding of categorical variables.built a loan default prediction model using scikit-learn and mlflow. the model consisted of two stages. the first stage was a feature selection module that filtered out non-relevant predictors based on the gini metric of a decision tree trained on the dataset. the second stage was a logistic regression model that predicted the loan status. this model helped minimize the risk of losing money while lending to customers., databases, pandas, industrial automation engineering, and huggingface to clean and transform text data into numerical data. the cleaning step involved several processes, improving deep neural, and scikit-learn to load, created feetgui, easy-to-use, regularization, python, making it highly portable and convenient., github, and portable computer vision-based tool that allowed users to estimate the effectiveness of epidural anesthesia by describing temperature changes in feet dermatomes, nltk, time series, m. tobon-henao, and process tabular data. the etl pipeline loaded the data from an s3 bucket and addressed issues such as missing values, ml frameworks, github actionsbuilt a two-stage food search engine using huggingface, publications, j. daza-castillo, chatgpt prompt engineering for developers, aws elastic file system, aws sdk for python (boto3), and save money while allocating resources more efficiently for the universidad tecnológica de pereira., aws efs, m. loaiza-arias, aws api gateway, visualization, mlflow, aws lambda, aws ec2, pydantic, natural language processing specialization, electronic engineer, and numpy to extract relevant and interpretable spatiotemporal biomarkers from a sorted collection of thermal images, streamlit, nlp, and spotifys annoy framework with the purpose of improving customer satisfaction, research assistant, b. lotero-londoño, and mlflow. the model aimed to classify newly created social media posts as either real or fake to prevent the harmful effects of their rapid propagation. the chosen transformer architecture enabled the model to make quick and accurate predictions in real-time, colombia, and tensorflow to transform eeg signals into spatiotemporal-frequency graphs. the new representation of the eeg signals helped to better understand how the brain works during motor imagery tasks and improved the performance of brain-computer interfaces in these tasks., and g. castellanos-dominguez., machine learning framework to support multi-channel time series classification in bci systems with preserved interpretability., developed a backend application using fastapi, education, pyqt5, aws, sqlite, aws ecr, raspberry pi, 202 - paper q1, springer international publishing, cartagena de indias, neuronal network and deep learning, aws s3, data, and germán castellanos-domínguez. subject-dependent artifact removal for enhancing motor imagery classifier performance under poor skills. sensors, brain-computer interfaces based on common spatial patterns for motor imagery recognition., computer vision |