Hi, I'm Mehul !

Lead Data Scientist at Grab, Singapore

I'm a product builder and lead analyst in a tech company. I prefer hands on execution and pair it with strong product sense with domain expertise in payments.

Currently Building

SQL Preview Extension, Data Analysis for cursor

The extension lets your LLM Agent run queries and analyse the data

SQL Preview Extension Icon View Extension
Profile picture for work section
Meili Mountains, Yunnan, China

Connect with me

LinkedIn Icon
Instagram Icon
Email Icon
Twitter/X Icon
YouTube Icon
GitHub Icon

Product Work

At the forefront of the "Builder PM" shift, I spearhead 0-to-1 launches of AI and data products at Grab, architecting full-stack prototypes from React frontends to agentic backends.

I've published a VS Code extension with 20K+ installs, and taken an internal AI analyst platform to adoption across Payments Analytics, where I also lead analytics strategy.

SQL Preview

VS Code Extension + MCP Server, Published on Open VSX

SQL Preview Extension Icon
  • Traction: 20K+ installs, a local-first SQL workspace for developers and the agents they work alongside.
  • Positioning: Local-first and read-only by default, the trust constraints enterprises need to adopt agents.
  • Scope: Own the full stack solo, product brief, 8-engine support, MCP tool design, docs and release.
VS Code Extension MCP Server TypeScript Local-First AI

Analytics Harness

Took a self-initiated prototype to a production grade product adopted across Payments Analytics. Leading a team of 8, I own product strategy, architecture and roadmap while remaining the primary developer.

  • Problem: Built a harness around the analytics process focusing on user personalisation and agent first ergonomics.
  • Tooling: Skills first harness with MCP connections to internal tools and custom analytics functions and workflows.
  • Evals: Implemented LLM-as-a-Judge and manual eval pipelines to iteratively improve agent accuracy.
  • Ecosystem: Skills marketplace and GitLab Pages publishing turned one-off analyses into reusable assets.
  • Security: Integrated OAuth to enforce RBAC, ensuring secure access to internal data ecosystems.
  • Adoption: Leadership demos, org-wide rollout, user onboarding drives, 100+ analyses published in H1.
Agentic AI MCP Integration LLM Evals Enterprise Security

Card Perks

PM and lead analyst for a 0 to 1 product through inception, feasibility, development, awaiting launch.

  • Role: Acting PM and lead dev for the 0-to-1 launch, validating feasibility via a self-built React MVP.
  • Tech: Engineered an LLM-powered pipeline to scrape and normalize unstructured offer data into a unified schema.
  • Product: Owned the 0-to-1 roadmap, scoping critical features to navigate strict resource constraints to launch.
0-to-1 Product React LLM Data Pipeline Payments & Offers