AI Systems Builder · Bangalore, India
I build and ship production AI systems, end-to-end.
Eighteen years as an operator and founder; since mid-2023 architecting and shipping production AI SaaS — retrieval, LLM orchestration, search, data, billing, and deployment. I prefer production-bound work over throwaway demos: systems that run under load, with every decision in them explainable.
Flagship — live in production
Vidhi — production AI research & drafting
A domain-specific AI research & drafting platform — RAG over a very large, citation-heavy corpus. Soft-launched with real users and live subscription billing — architected, built, and operated end to end.
Production RAG platform
Fast, source-cited answers over 750,000+ document chunks — with the full product built around the model, not just an API wrapper.
The platform
A shared platform that ships production AI fast
Vidhi runs on a multi-tenant platform I architected to take AI products from idea to production quickly — auth, billing, RAG, chat, search, and LLM orchestration, hardened once and reused. A sample of the systems it runs:
Deterministic classification & scoring engine
Rule-based, with a full auditable decision trail — chosen over an LLM where traceability and defensibility matter more than flexibility.
Multi-stage generative-media pipeline
Brief → storyboard → asset generation → voiceover → render, orchestrated through resilient background workers across several media providers.
Entity & relationship knowledge graph
A graph store layered over Postgres for entity-and-relationship intelligence, kept in sync from the primary system of record.
Resilient ingestion & extraction
Browser-automation ingestion for dynamic, JavaScript-heavy sources, plus multi-format document extraction/OCR and normalization.
Shared multi-tenant core
Unified auth, billing, RAG, chat, search, and content — so each new product ships as a thin shell, not a rebuild. Improvements compound across every app.
Architecture walkthroughs under NDA
Under mutual NDA I walk my own proprietary systems — real screenshots, workflows, architecture. I never use client work as public proof.
What I build
Systems I build
Production RAG & document intelligence
Hybrid retrieval, source-cited answers, evaluation harnesses, and anti-hallucination controls that hold up in production.
Multi-provider LLM orchestration
Routing, fallback, circuit breakers, cost/token governance, and prompt caching across several model vendors — no single-vendor lock-in.
AI SaaS architecture & multi-tenancy
Auth, org tenancy, billing, and the platform plumbing that turns an AI capability into a real, operable product.
Search, ingestion & evaluation
Hybrid lexical + vector search, resilient ingestion and extraction pipelines, and measurable quality harnesses.
Launch & operations infrastructure
Deployment, observability, disaster recovery, and the operating discipline to run systems under real load.
Before the code — 18 years operating
Operator & founder track record
MITx MicroMasters, Statistics & Data Science (MIT) · Independent Directors Certification, IIM Bangalore · MBA · B.Tech, Computer Science
Let's talk
Vidhi, or a hard AI problem
Whether you're interested in Vidhi, or your team is facing a serious production-AI, RAG, or architecture challenge — I'm glad to talk.