Skip to content
View andem-divya's full-sized avatar

Highlights

  • Pro

Block or report andem-divya

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
andem-divya/README.md

Hi, I'm Divya 👋

Software Engineer | Data & AI Engineer · Fremont, CA

🎓 MS Applied Data Science (AI concentration) · University of Michigan, May 2026
🎓 B.Tech Electrical Engineering · IIT Hyderabad (Indian Institute of Technology)


Tech stack

Languages & data Python · SQL · PySpark · Pandas · HTML/CSS/JS
Cloud & infrastructure AWS (Lambda · S3 · RDS · SQS · API Gateway · Bedrock) · Docker · CI/CD
AI & GenAI LangChain · RAG · VectorDB · AWS Bedrock · Prompt Engineering
ML Scikit-Learn · PyTorch · XGBoost · Random Forest · KMeans · PCA · NLP
APIs & backend Flask · FastAPI · SQLAlchemy · RESTful APIs · Microservices


Featured projects

Project What it does Stack
MoveSmart End-to-end city intelligence platform — ingests 5 government datasets (Census, FBI, CDC, NOAA, EPA), engineers features, applies PCA + KMeans clustering and content-based filtering to score 900+ U.S. metros across 6 life themes. RAG pipeline with ChromaDB embeddings and semantic search generates personalized AWS Bedrock LLM summaries. Fully deployed. Python · PySpark · ChromaDB · RAG · LangChain · AWS Bedrock · Streamlit
Airbnb Success Prediction What makes an Airbnb listing successful? Predicts review activity across 52K California listings using XGBoost, Random Forest, KNN and Linear Regression, enriched with FBI crime, EPA walkability, and Census data. Unsupervised segmentation via PCA + KMeans reveals 4 distinct listing archetypes by property scale, stay strategy, and neighborhood context. Python · XGBoost · Random Forest · KMeans · PCA · GeoPandas · Scikit-Learn · Pandas
US Automotive Trade Analysis How do GDP and tariffs shape U.S. automotive trade? Analyzes imports, exports and trade balance across 30+ countries from 2008–2022, integrating ITA trade data, World Bank GDP, and WITS tariff datasets. Includes Spearman correlation, OLS regression, and animated visualizations revealing how the 2018 tariff spike and COVID-19 impacted trade flows. Python · Pandas · Matplotlib · Seaborn · Scipy · OLS Regression

Experience highlights

  • Built serverless REST APIs on AWS Lambda + API Gateway, containerized with Docker, deployed via GitLab CI/CD — exposing golden records for insurance workflows at Accenture
  • Designed event-driven microservices with AWS SQS processing real-time insurance domain events for transactional downstream systems
  • Automated a 4M+ row Excel process into a Python pipeline — cut processing time from half a day to under 10 minutes at MatrixIntelligence
  • Built end-to-end ETL pipelines delivering structured datasets to S3 for downstream ML systems, saving 20+ hours/month of manual effort
  • Built an HR data agent with LangChain and custom tools enabling natural language queries over employee datasets via prompt engineering and LLM pipelines

Currently


"Build things that work. Then make them work better."

Pinned Loading

  1. movesmart movesmart Public

    Find the city that fits your life. MoveSmart analyzes economic, social and environmental indicators across 900 U.S. metros and matches you to places that align with what you actually value, with AI…

    Jupyter Notebook

  2. Airbnb-Success Airbnb-Success Public

    What makes an Airbnb listing successful? Predicting review activity and segmenting 52K California listings enriched with crime, walkability and census data.

    Jupyter Notebook

  3. US-Automotive-Trade-Analysis US-Automotive-Trade-Analysis Public

    How do GDP and tariffs shape U.S. automotive trade? Analysis of imports, exports and trade balance across 30+ countries from 2008–2022 using ITA, World Bank and WITS data.

    Jupyter Notebook