| What | How |
|---|---|
| Build reliable data pipelines | Batch and streaming ingestion with Airflow, PySpark, dbt and Kafka on AWS, Azure or Google Cloud |
| Design fast, low cost lakehouses | Models on Databricks, Microsoft Fabric, Snowflake and BigQuery that are quick to query and cheap to run |
| Put AI on top of your data | RAG search, LangGraph agents and MCP servers that answer questions from your own data |
| Ship it to production | Data quality checks, Docker, Kubernetes, Terraform and CI/CD so it keeps working after launch |
AI Data Engineer with 5 years of experience across retail, insurance and customer analytics. Currently building the GCP catalog data platform at Walmart. M.S. in Applied Computer Science from Grand Valley State University. Based in Grand Rapids, MI.
| Project | What it does | Stack |
|---|---|---|
| Flight-Data-Pipeline | Airflow pipeline every 30 minutes moving live OpenSky data through Bronze, Silver, and Gold layers into Snowflake; keyed MERGEs prevent duplicates on reruns | Python · Airflow · Snowflake |
| AWS-Cost-Optimizer | Streamlit dashboard combining AWS billing and inventory across 7 resource categories, with a LangGraph advisor that suggests savings | Streamlit · LangGraph · Terraform · AWS |
| NYC-Green-Taxi | Azure Synapse pipeline processing 2.3M trips across 19 monthly partitions into a Power BI report on demand and payments | Azure Synapse · Power BI |
| Bing-News-Intelligence | PySpark notebooks curate articles and score sentiment, with Delta tables feeding a Power BI report and negative news alerts | Microsoft Fabric · PySpark · Delta · Power BI |
| Crop-Disease-Management | Relational database backend with alerts, recommendations, and yield tracking | MySQL · Stored Procedures |
| Project | What it does | Stack |
|---|---|---|
| Advanced-RAG-Agent | Conversational RAG agent with an 8 node LangGraph workflow: query reformulation, topic validation, relevance filtering, and bounded retries | LangGraph · RAG · Python |
| HRMS-MCP-Server | MCP server exposing 12 HR tools (employee records, leave, meetings, tickets, email) for Claude Desktop and other MCP clients | FastMCP · Python · Pydantic |
| Blog-Agent | LangGraph agent using GPT-4o-mini to create SEO optimized blogs in 6 languages, deployed on Vercel | LangGraph · OpenAI · FastAPI |
| E-Commerce-Chatbot | Chatbot with semantic routing, vector FAQ retrieval, and natural language product search | Groq LLM · ChromaDB · Streamlit |
| Badge | Certification | Skills |
|---|---|---|
| SQL AI Developer Associate (DP-800) Sep 2026 |
T-SQL and database design, vector, full-text and hybrid search, embeddings and RAG inside SQL, row-level security and encryption, CI/CD with SQL Database Projects | |
| Azure AI Apps and Agents Developer Associate (AI-103) Aug 2026 |
Generative AI apps and agents in Microsoft Foundry, retrieval and grounding pipelines, document extraction, vision, speech and text analysis, responsible AI and monitoring | |
| Fabric Analytics Engineer Associate (DP-600) Jan 2026 |
Lakehouse and warehouse design, data preparation with SQL and PySpark, semantic models and DAX, security and deployment of analytics solutions | |
| Fabric Data Engineer Associate (DP-700) Oct 2025 |
Batch and streaming ingestion, transformation with PySpark and SQL, orchestration with pipelines, monitoring and performance tuning | |
| Databricks Data Engineer Associate May 2025 |
Lakehouse and Delta Lake, ELT with Spark SQL and PySpark, incremental loads with Auto Loader, production jobs and workflows, Unity Catalog governance |
- Co-author, Soil Moisture Prediction and Benchmark Reliability: A Review with Multi-Site SMAP–ISMN Analysis. International Journal on Advances in Intelligent Systems, vol. 19, 2026. Read the paper
- Graduation Regalia Scholarship, GVSU, for outstanding academic achievement (Winter 2026)
- Pratibha Award, Government of Andhra Pradesh, for excellence in the Secondary School Certificate exam (2017)
All certificates: Google Drive folder
Have a data or AI problem to solve? Reach me at puneethkumaramudala@gmail.com







