Hi, I’m Subodh Pandey 👋

Hi, I’m Subodh Pandey 👋

I am a Backend Engineer with 8+ years of experience specializing in data-intensive applications, distributed computing (PySpark/Databricks), and API-driven architectures. Currently, my focus is bridging the gap between big data infrastructure and generative AI by building scalable agentic workflows and production-ready RAG systems.

🚀 What I’m Currently Working On

• Agentic Workflows: Orchestrating multi-step LLM reasoning agents (Text-to-SQL, Data Analysis) using Python.
• Backend Serving: Building highly concurrent, low-latency APIs using FastAPI to serve foundation models.
• Model Optimization: Quantizing Transformer models and converting them to ONNX for edge/production deployment.

🛠️ Technical Stack

• Backend & APIs: Python, FastAPI, RESTful Design, Docker
• Data & Distributed Systems: PySpark, Databricks, SQL, PL/SQL
• AI/ML Ecosystem: LangChain, Hugging Face, FAISS, ChromaDB, ONNX
• Cloud & MLOps: AWS (Bedrock, SageMaker), CI/CD pipelines

📈 Highlighted Projects (Pin these repositories below)

1. [Insert Repo Link: enterprise-rag-fastapi]: A containerized backend utilizing FastAPI, ChromaDB, and LLMs for semantic search over enterprise documents.
2. [Insert Repo Link: text-to-sql-agent]: A multi-modal AI agent that translates natural language queries into complex SQL schemas for database querying.
3. [Insert Repo Link: transformer-onnx-optimization]: Benchmarking latency and payload size improvements by quantizing a text classification model for production environments.