Jay Prakash

I’m a Full-Stack Python Engineer with 3+ years building production-grade backend systems in FastAPI and Django. I specialize in LLM-powered applications and Retrieval-Augmented Generation (RAG) pipelines for high-volume document processing and analysis.

I build LangChain-based RAG pipelines on PostgreSQL with pgvector, enabling semantic search and retrieval across large, unstructured document corpora. I integrate OpenAI and Anthropic APIs to automate summarization, information extraction, and decision-support in document intelligence workflows.

I integrate third-party and internal APIs (CRMs, databases, webhooks, and LLM services) and build data pipelines that replace manual document- and case-handling steps with automated workflows. I own architecture decisions end-to-end—from data modeling (PostgreSQL, MongoDB) to secure auth (JWT, OAuth2)—optimizing for scalability and performance.

I’m comfortable in fast-paced, cross-time-zone startup environments and collaborate closely with founders and cross-functional teams to ship features that move product and business outcomes. I also mentor junior engineers on AI-assisted best practices, and I keep learning through AI-native projects, automation tools, and agentic workflows.