• Assisted and reviewed in development of a hash table project to be successfully deployed to 350+ students.
• Enhanced students’ algorithmic programming skills by hosting one-on-one guidance through weekly office hours.
• Graded project solutions to assess runtime/space complexity, ensuring optimal algorithmic efficiency.
• Spearheaded the development of a full-stack web application Proof-Of-Concept integrating with existing data ingestion processes for a Dashboard UI, enhancing the system observability for an internal team of 10+ engineers.
• Migrated web application to a cloud native and serverless architecture, enabling live querying from a DynamoDB database and real-time metrics of data loads into S3 buckets.
• Built multiple highly reliable and performance-driven REST APIs that efficiently handle GET and POST requests to query and store data in a DynamoDB database.
• Demonstrated Proof-Of-Concept web application across cross-functional teams, receiving positive feedback and generating significant interest, and effectively convincing them of the feasibility of the dashboard concept.
• Awarded Most Innovative Hack in company-wide hackathon using the MERN technology stack.
• Developed a program to preprocess text data in over 1500 contracts with an accuracy of 95% through a multi-layered approach using NLTK for further analysis with machine learning models
• Led a team of 3 to create detailed reports of vendor data using pandas and NumPy to leverage statistical analyses, resulting in improved decision-making and high-performance vendor evaluations
• Conducted geospatial data analysis with the GeoPandas library to identify geographically disadvantaged areas, utilizing third-party API endpoints and presenting results to stakeholders, contributing evidence-based decision making.
• Automated functionality of downloading targeted contracts by developing a Python script that locates identified contract IDs in an Excel workbook, reducing the turnaround time significantly.
• Utilized Natural Language Processing technology to remove potential stop words and use word tokenization to feed machine learning models cleaned data.
• Collaborated with development team under Agile development cycle to conduct User Acceptance Testing (UAT) on user stories, identifying up to 1-2 bugs daily, saving significant time spent on rework.