Meet Omar Sanchez!

• ● CQPA: 4.12/4.0
• ● Coursework: 11-785 Introduction to Deep Learning, 10-601 Introduction to Machine Learning, 11-667 Large Language Models, 95-828 Machine Learning for Problem Solving, 95-865 Unstructured Data Analytics.
• ● Worked as a Teaching Assistant (TA) for the courses Unstructured Data Analytics for two semesters and Object Oriented Programming with Java for 1 semester. Served as an officer at the Colombian Student Association.

● CQPA: 4.4/5.0

• ● CQPA: 4.5/5.0
• ● Served on the team for the robotics football world championship “Robocup” in 2019 that reached semi-finalist status in the “Shield Category”. Developed a path planning algorithm and main behaviors for the striker role.

• ● Designed and developed an AWS-based batch NLP service for cleaning, storing, processing, and aggregating open-ended questions using AWS S3, AWS Athena, AWS Step Functions, AWS Lambda, and AWS Comprehend while employing only one-third of the time allotted for this.
• ● Established a full CI/CD pipeline with unit and integration testing, for the deployment of the solution using AWS CloudFormation, PyTest, and internal DevOps tool.
• ● Integrated the NLP service with a third-party API, serving as the user interface for enhanced UX.

• ● Retrieved data from REST APIs to generate automated reports using Python, AWS Lambda, Docker, AWS S3, AWS DynamoDB, AWS SNS, and AWS Step Functions. Deployed using Serverless Framework.
• ● Designed and implemented a layered data lake using AWS S3, the AWS Glue data catalog, and PySpark.
• ● Developed a pipeline to extract data from both PostgreSQL and S3 to create a data warehouse following Kimball’s modeling technique in AWS Redshift using Python, Airflow on AWS EC2, AWS Glue, Apache Spark, DBT, and SQL.
• ● Implemented CI/CD pipelines to automate infrastructure deployment using GitHub Actions and Terraform.
• ● Consumed and transformed streaming data to produce aggregated statistics, store data, and create real-time dashboards using Python, AWS Kinesis, AWS S3, and Splunk.

• ● Built, tested, and documented several data pipelines using Apache Spark, Control-M, and other proprietary big data tools including Datio for an international bank.
• ● Deployed hundreds of Control-M data workflows and automated the creation of workflow-definition XML files using Python, RegEx, and Pandas, decreasing by a factor of 4 the time-to-production of new pipelines.
• ● Ensured data quality and enforced rules of completeness, consistency, integrity, etc. according to data governance expectations.

• ● Built a tool using Selenium Web Driver, Python, Bash, and Docker to automatically execute tests to evaluate the impact of change on risk metrics, making the whole process four times faster.
• ● Led and orchestrated deployments from development to production using Jenkins and Bitbucket.

• ● Designed, built, and maintained single-handedly and proactively a GCP-hosted Bigquery data warehouse to support complex operation reports, accounting reports, and ad hoc analysis for an online pay tech company in Latin America. The data warehouse enabled a holistic view of a range of data sources (i.e. MySQL, Google Datastore, Google Cloud Logging, CSV files, spreadsheets, Neo4J/API). Leveraged the power of GCP (Cloud functions, Dataflow/Apache Beam, Dataprep, Bigquery, Google Cloud Storage, Pub/Sub ), Data Build Tool (DBT), Javascript, and Python. This helped the business fulfill several data needs and surface costly mistakes that required attention by the C-suite.
• ● Designed, built, and maintained various analytical reports using SQL and Data Studio (now Looker) oriented to C-level executives.
• ● Implemented integrations between Google Bigquery and Spreadsheets to serve different reports and tables to provide a familiar and friendly interface to the warehouse to non-technical users.
• ● Developed several tools using Javascript and Node.js to automate and optimize operative processes decreasing by a factor of 10 the errors to be manually checked and fixed.
• ● Identified, analyzed, and resolved complex errors within the service using SQL and Bigquery, leveraging a deep understanding of the REST API services and synchronous/asynchronous messaging styles.
• ● Automated several REST API tests using Postman and Javascript, decreasing the testing time to just one-fifth.