Suyog Shah
Immediate joiner

AI-First Engineer

Suyog Shah

I build full-stack products from scratch: cloud backends, web apps, mobile apps, and the AI that connects them.

8 years experiencePune, IndiaOpen to senior roles
AngularNode.jsJava
  • Multi-tenant SaaS
  • AI orchestration
  • Cloud-native on GCP
  • Web + Mobile + Backend

About

I'm a full-stack engineer with 8 years of experience shipping production software — most of it for marketing and healthcare teams that need to move fast without breaking. I started writing Angular and Firebase apps in 2017 and have spent the last few years migrating that world over to Next.js, TypeScript, Fastify, Supabase, and Google Cloud.

Right now I spend most of my time building AI-first systems — orchestrating Claude, OpenAI, and Gemini inside multi-tenant SaaS platforms, and figuring out how to make that orchestration cheap, reliable, and easy to operate.

Outside work, I'm usually [your non-work fact — cycling around Pune / reading sci-fi / playing chess / cooking].

Technical Toolkit

The tools I build with, grouped by layer.

Languages

TypeScriptJavaScriptPythonDartJava

Frontend & Mobile

Next.jsReactAngularTailwind CSSFlutter

Backend

Node.jsFastifyExpress

Data

SupabasePostgresFirestore

Cloud

GCPFirebaseVercel

AI / LLM

ClaudeOpenAIGemini

Tooling

DockerGitHub ActionsVitestPino

How I Work

How I take a problem from idea to deployed, and back.

  1. 01

    Understand

    Start with the actual problem. Talk to the team or client to understand what they really need, not just what they asked for.

  2. 02

    Prototype

    Build a quick POC to validate the approach. Small enough to throw away if needed, real enough to prove the idea works.

  3. 03

    Build

    Iterative implementation, end to end first. Get every layer working at a basic level before polishing any single layer.

  4. 04

    Ship

    Deploy to production with monitoring and rollback in place. Reliability starts at shipping, not after.

  5. 05

    Operate

    Watch real usage. Fix what breaks. Learn what users actually do versus what we expected.

  6. 06

    Iterate

    Feed learnings back into the next problem. The cycle restarts with better context every time.

...and back to step 01 with the next problem.

Contact

Send a message and I'll get back to you within a day or two.