Prev @ Multibagg AI · National Finalist IFF-FinTech Olympiad’24 · IIT Patna'27
Working on AI Agents, Quant and Backend
Founder's Office & AI Engineer
PineconeQdrantRedis
BullMQGrafanaAzureDockerJan 2025 - Jun 2026 · 1 yr 5 mos
Worked directly with the founder to build and scale production AI systems across agent orchestration, retrieval, evaluation, financial data and user-facing product workflows.
Bachelor of Science
Aug 2023 - May 2027 · 3 yrs 9 mos
Major: Computer Science and Engineering
Minor: Data Science and Artificial Intelligence
Work in progress
LangSmithPostgreSQLRedisPineconeQdrantReinforcement learningRunnable city foundation · cloud agent, memory, and learning in progress
Imagine Clash of Clans for your codebase: every file is a village building, and you can watch builders work on it. I'm building a cloud agent to fix issues and open PRs, with six layers of memory and a reinforcement learning loop planned to make its decisions more accurate and cost-efficient.
Razorpay AI Buildathon · verified Test Mode recovery flow
An explainable revenue-recovery system for Razorpay merchants: AI proposes one bounded action, deterministic policy guards execution, and durable workflows follow failed payments to auditable outcomes.
Make recovery predictable, explainable, and trustworthy.
Frontend · Nifty 50 paper-trading arena
An experiment where LLMs paper-trade Nifty 50 stocks using market news, including each story's sentiment and importance, custom technical indicators, quantitative analysis, and portfolio strategies. The read-only dashboard lets you follow their trades, rankings, and portfolio performance, with reproducible Redis-backed replay.
BullMQPineconeAzurePostgreSQLRanked financial news · traceable events · Indian equities first
A financial-news platform for following topics and regions on demand. It fetches news, filings, and disclosures, ranks their importance, and connects related items to structured, traceable events with source context and materiality. Its current focus is Indian equities, helping researchers move from a headline to the evidence behind it.
From research to rigorous, reproducible execution.
Room AI take-home · geometry owns centimetres, the LLM cannot write wall lengths
Imagine assessing a house for an insurance claim: every wall and patch of damage needs measuring. This local pipeline uses iPhone LiDAR to map the floor area and room geometry, then computer vision and LLMs to find and classify damage, producing a measured floor plan with evidence for an estimate.
Replacement-cost survey · models propose, they don't write count or money
An iPhone LiDAR app uses a live AI guide to walk a surveyor through the library's floor plan, damage, shelves, and other objects. Computer vision counts books from their spines; ISBN, barcode, and title evidence help identify editions; web searches draft local prices. After sealing, two models independently review the same evidence, and Jev suggests recapture or human review under fixed rules. An offline reinforcement-learning loop is designed to improve that routing, while a signed report estimates the whole library's replacement cost.
Evidence-first analysis · reports, transcripts and frame sampling
A multi-agent analysis pipeline that ranks public Instagram posts, extracts video and audio evidence, and turns measurable creative patterns into an adaptable strategy report.
Automate the boring, ship more confidently.
Workshop map
How the pieces connect to build and ship things.
follow the build →
what users see and interact with
the contracts and gateways
store, shape and access
work that keeps moving after the click
reach users and ship reliably
models, agents, data and retrieval
the things that help me test, observe and ship
Mine The Model · Celesta IIT PatnaStock-price ML contest by NJack ML IIT Patna & Cynaptics IIT Indore — beat the benchmark.
6-week Elementary & Advanced quant finance programme by Quant Club, IIT Kharagpur.
Tourist's city foundation is running. Here is the build plan for the cloud coding agent, scoped memory, dynamic context, GitHub delivery, and learning that should follow.
From a 20-minute delivery estimate to LLM latency: explore how a line learns, why we square its mistakes, and when a good-looking prediction stops being useful.
GPT-6 Astra is not just a stronger reasoning model. Its computer-use gains, critical cyber capability, and monitorability limits change how we should design and supervise AI agents.
