Skip to content

Rajdeep Mondal

Currently an AI engineer. I take AI systems from a demo that works to one that scales in production.

Generalist: startups, software, books, gym, trekking.

Currently

Build
Autonomous Research Agent for drug target discovery at Elucidata.
Read
Money for Couples, by Ramit Sethi.
Train
Preparing for Hyrox.
Rajdeep Mondal
  1. Senior Data Scientist, GenAI Lead, Elucidata

    Jan 2024-Present

    I lead all GenAI initiatives for autonomous research in drug target discovery and validation.

  2. Data Scientist, Elucidata

    Jul 2022-Dec 2023

    I took generative AI at Elucidata from first prototypes to shipped systems across four business verticals.

  • PMCGrab (opens in a new tab)

    Extracts PubMed Central papers into structured JSON: sections, tables, figures, references.

  • ACE AI (opens in a new tab)

    AI engineering interview prep with a 2,049-question bank across 28 topics, built around study, quiz, review, and progress loops.

  • Oncopacket (opens in a new tab)

    A Python package that converts cancer data into GA4GH phenopackets, so records from different studies can be read by the same tools.

  • Vaidya (opens in a new tab)

    Voice-first navigator for Indian public health schemes. A caller asks in their own language and hears which coverage they can claim.

  • Caliper (opens in a new tab)

    Local-first analytics for AI coding spend. It reads Codex CLI and Claude Code logs on disk, prices usage at API rates, and breaks cost down by project, PR, model, and session.

  • WHOOP MCP Server (opens in a new tab)

    Read-only MCP server that connects WHOOP records to Claude, ChatGPT, and other MCP clients so health and training questions are answered from computed data.

One specific thing gets the first reply: a wrong claim, a broken step, a stuck build.