Jaspal B.

I build AI features and web apps that ship on time and stay stable in production. Product teams and founders bring me in to add RAG search and ship clean Python APIs. I also tighten up React front ends. I've spent 12.8 years delivering systems under real traffic. React with TypeScript on the client. Python on the server. REST APIs with clear docs and versioning. AWS for deploys with Docker and Kubernetes. CI/CD keeps releases repeatable. PostgreSQL or MongoDB, chosen for the workload. Indexes and schema patterns that keep queries fast as data grows. My AI work centers on LLM apps and RAG. I use LangChain for orchestration and OpenAI models where they fit. I design retrieval, chunking, embeddings, and caching tuned for latency and answer quality. Token spend stays in check. I fold in auth and rate limits. Metrics are tracked from the start. What I can help with: - Ship a focused MVP fast, React on the front. Python REST on the back. - RAG features that answer with context. LangChain plus OpenAI with eval and guardrails. - Data heavy dashboards and CRMs. Real time analytics backed by clear APIs. - AWS deployments with Docker and Kubernetes. CI/CD that ships with one click. - PostgreSQL or MongoDB schema design and indexing. Query tuning when performance slips. How I work: - Short discovery call. We lock scope and key flows in a shared doc. Success criteria are clear. - Prototype or API spec in 48 to 72 hours, when the brief allows. - Milestones with weekly demos. You see progress in a hosted environment. - Clean Git history. Readme and handover docs at the end. I'm available 20 to 30 hrs per week, IST. I can overlap with US and EU hours. Typical work, add an AI feature to an existing app. A 10 day RAG proof of concept. Migrate a core module to a Python microservice without downtime.