Skip to content
View parbhatkapila4's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report parbhatkapila4

Report abuse

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

Report abuse
parbhatkapila4/README.md

Parbhat Kapila

Full-stack engineer specializing in production AI infrastructure. I build RAG systems, semantic search platforms, and real-time data pipelines that handle thousands of documents with sub-2-second response times.

What I've Shipped

Have made production AI applications serving real users:

  • Document intelligence processing 10,000+ documents at 99.9% uptime
  • Semantic search with hybrid BM25 + vector similarity across 200+ code repositories
  • Real-time email indexing with multi-LLM orchestration and automatic fallback (OpenAI → Gemini → Claude)

Engineering challenges I solve:

  • RAG architecture with pgvector and intelligent chunking strategies
  • Multi-provider API orchestration with fallback handling and rate limit management
  • Real-time WebSocket systems with Redis caching, achieving sub-2s query responses
  • Type-safe full-stack applications with tRPC and end-to-end TypeScript
  • CI/CD pipelines on AWS/Vercel with Docker containerization

Technical Depth

AI & Backend
RAG Architecture • LangChain • OpenAI API • pgvector • Vector Embeddings • Semantic Search
Python (FastAPI) • Node.js • PostgreSQL • Redis • Multi-LLM Orchestration

Frontend & Full-Stack
TypeScript • React • Next.js • tRPC • WebSockets • Real-time Systems

Infrastructure
AWS (EC2, S3, Lambda) • Docker • Vercel Edge Runtime • CI/CD • GitHub Actions

Integrations
OAuth (Gmail API) • Payment Processing (Razorpay/Stripe) • Third-party API orchestration

Track Record

  • 500+ commits in 2025 (consistent shipping velocity)
  • 99.9% uptime across production systems
  • Sub-2 second response times at scale
  • Processing 10K+ documents in production
  • 80% reduction in developer onboarding time with AI documentation tools

Currently

Open to full-time remote roles at startups where I can own technical architecture and ship features that matter. Available with US/EU timezone flexibility.

Connect

Portfolio LinkedIn Twitter Email


Looking for an engineer who can architect AI systems, integrate complex APIs, and ship production features fast? Let's talk.

Pinned Loading

  1. Visura Visura Public

    An Enterprise AI Document Intelligence Platform Production SaaS processing 10K+ documents with RAG, multi-LLM orchestration, real-time streaming, and enterprise billing. Sub-2s response times, 99.9…

    TypeScript 1

  2. RepoDocs RepoDocs Public

    An AI-Powered Code Documentation Platform Automated documentation engine with hybrid search (BM25 + vector). Transformed 200+ repos into queryable knowledge bases, cutting onboarding time 80%

    TypeScript 2

  3. Vector-Mail Vector-Mail Public

    An AI Email Intelligence Platform Real-time email intelligence with multi-provider AI fallback, semantic search, OAuth integration. Handles incremental sync and streaming with 70% cold start reduct…

    TypeScript 2

  4. ClimaSphere ClimaSphere Public

    A Production weather dashboard with real-time data visualization, intelligent caching strategies, and TypeScript-first architecture. Showcases scalable React patterns and performance optimization.

    TypeScript