✦ CLICK THE SKY
--:--

VEDANSH KUMAR_

DEVELOPER · SOLUTION ARCHITECT · GENERATIVE AI

₹2cr+revenue shipped
$100K+per deployment
500K+monthly transactions
5,000+users served
PRESS START ↵

SKIP INTRO → recruiter overview

OVERVIEW

The 30-second version, for people in a hurry.

Developer · Solution Architect · Generative AI — LLM systems, document intelligence, BFSI ML, games

I build production GenAI pipelines, document-intelligence frameworks, and BFSI ML systems — work that has generated ₹2 crores+ in revenue, powers $100K+ per-deployment products, and serves 5,000+ users. Currently building autonomous agents at Sword; before that, four years of enterprise AI at Xebia.

★ FEATURED STAGES

The flagship work. Every card opens a full case study.

★★★ Jun 2026 – Present

Sortie — Job-Application Autopilot

A self-hosted agent that applies to jobs for you inside your own logged-in browser, tailors the résumé per role without inventing facts, and keeps the ledger so you always know where you stand.

Browser AutomationPlaywrightFastAPINext.js
▸ ENTER STAGE
★★★ Jun 2026 – Jul 2026

Idolwild — An Ant-Farm for LLM Minds

A browser god-game where 60 stone-age agents perceive, remember, plan, gossip, and invent religions — a deterministic cognition engine that runs for free, with an LLM rationed in as the smallest, most-replaceable part.

TypeScriptthree.jsVercel AI SDKZod
▸ ENTER STAGE
★★★ Jun 2026 – Present

Greybeard — Senior Judgment for Coding Assistants

Playbooks and tripwires distilled from real work across ML, quant, agents, RAG and backend, so any coding assistant stops making the junior mistakes it was always going to make.

LLM PlaybooksAgent SkillsKnowledge DistillationAdversarial Review
▸ ENTER STAGE
★★★ Jan 2026 – Feb 2026

AutoTrader — Regime-Aware Quant ML Lab

A two-tier quant-ML lab for NSE equities — 36 regime-specialist models under a GPT-based "checker" (a Decision Transformer + RL ensemble), with honest walk-forward backtesting and a documented win for simplicity.

PythonXGBoostLightGBMPyTorch
▸ ENTER STAGE
★★★ Sep 2025 – Mar 2026

Neural Core — Retrieval as a Control Problem

A retrieval-and-reasoning backbone that treats RAG as a control loop — grade, verify, reflect — over an agent-agnostic blackboard where new capabilities plug in with zero rewiring.

PythonFastAPICorrective RAGSelf-RAG
▸ ENTER STAGE
★★★ Jan 2026 – Jan 2026

Chomper — One Document Core, Three Surfaces

One document-processing core that chomps 36+ file formats, fronted by a library, a CLI and an MCP server — with a swappable four-stage pipeline and a token-optimized wire format.

PythonMCPDocument ParsingSemantic Chunking
▸ ENTER STAGE
★★★ Jan 2024 – Dec 2024

XParser - Enterprise Document Intelligence

Enterprise-grade document parser sold at $100K+ per deployment to healthcare clients in USA and Europe

Enterprise AIDocument ParsingOCRLLM
▸ ENTER STAGE
★★★ Jan 2023 – Dec 2024

BFSI ML Systems — Transaction Intelligence, Underwriting & Fraud

Production ML systems for enterprise banking — transaction intelligence, fraud, underwriting, and next-best-action — delivered for enterprise banking clients across UK and Indian markets.

Machine LearningBFSITransaction CategorizationFraud Detection
▸ ENTER STAGE
RESUME.PDF — ACQUIRE ITEM ↓ ALL 43 STAGES — ENTER LEVEL SELECT →