Aman Kumar Chhari

Aman Kumar Chhari

Applied AI Engineer
& full-stack

I ship AI features inside real products, with the guardrails that make them safe to run.

available immediately · open to relocate
  • Next.js
  • React
  • TypeScript
  • Node.js
  • Postgres
  • Claude API
  • Resume
  • GitHub
  • LinkedIn
  • Email

about

Hi there.

Applied AI engineer who ships AI features inside real products, with guardrails in place and JMeter load tests that show they hold under real concurrency. I work alone and build end to end: system design, typed backend, UI, payments and deployment. Freelance since 2023; B.Tech CSE 2026.

Recent work: Stack, CloseBy, ScanCart.ai and Orderly.

Hackathon winner (Reimagine Web Dev Hackathon, Component Making Hackathon), with national-level finishes at Hacksagon 2025, HackGSon Hackathon and Smart India Hackathon. Every entry solo.

B.Tech, Computer Science & Engineering, Jaypee University of Engineering & Technology, Guna (2022 – 2026).

GitHub·Résumé·amankumarchhari@gmail.com

Based in
Guna, Madhya Pradesh
Location
Open to relocate
Start
Immediately
Education
B.Tech CSE, 2026
Looking for
Applied AI engineerFull-stack engineerSoftware engineer (SDE-1)

applied ai

What is applied AI?

Applied AI is the engineering that puts language models to work inside real products: deciding where a model actually helps, connecting it to the app's data and tools, checking what it returns, and keeping cost and failures under control. The model is one part of the system, not the whole product.

Why an applied AI engineer?

A model demo takes an afternoon. Making it dependable for real users, with real data and real money, is the hard part. That is the job of an applied AI engineer, and this is how I approach it:

  • Guardrails first

    Every model call passes one gateway with auth, rate limits, a kill switch and a daily spend cap.

  • AI drafts, people decide

    The model proposes stock imports, replies and restock ideas. Nothing is saved until the owner approves it.

  • Checked output

    Responses are validated against a Zod schema. Malformed output fails closed: the user sees an error and nothing is saved.

  • Measured, not guessed

    Eval cases cover misspellings, Hindi aliases and gibberish input, and JMeter load tests check the guardrails under concurrency: 50 buyers racing for the last 5 units still make exactly 5 orders.

work

Things I’ve shipped.

Stack: A local-first reading app with spaced review of highlights

Stack · A local-first reading app with spaced review of highlights

More on GitHub

experience

Experience & wins.

Experience

  • Freelance Full-Stack & AI Developer

    Independent · Guna, MP (Remote) · Feb 2023 – Present

    2 client projects

    • Rent It, vehicle rental platform · Aug 2025code

      React and Tailwind app where customers browse vehicles, book and pay through Razorpay, with Firebase sign-in, protected routes, a map section and booking history.

    • Gurudev Public School, Guna · Oct 2025code

      The school's 15-page website in React, Vite and Tailwind: academics, facilities, gallery, news, director's and principal's desk, holiday homework and the public disclosure page.

  • Web Development Intern

    Zidio Development · Remote · Aug – Sep 2024

Hackathons · all entered solo

  • Finalist (600+ teams), Hacksagon 2025, ABV-IIITM Gwalior, as team MANADcertificate
  • Selected pitch, Rajasthan AI Builders Pitch-a-Thon 2026 (National Conference on e-Governance), Jaipur, with ScanCart.aicertificate
  • National Finalist, HackGSon Hackathon
  • Semi-Finalist, Smart India Hackathon
  • Winner, Reimagine Web Dev Hackathon
  • Winner, Component Making Hackathon
  • 7th Rank, Skill-A-Thon 2024

skills

Tools of the trade.

  • Next.js

    used in Stack, CloseBy

  • React

    used in ScanCart, Orderly, Rent It

  • TypeScript

    used in Stack, CloseBy

  • Node.js

    used in ScanCart, Orderly

  • Postgres

    used in CloseBy, Orderly

  • Claude API

    used in CloseBy, ScanCart, Stack

Frontend

Next.js · React · TypeScript · Tailwind CSS

Backend

Node.js · Express · FastAPI · Python · Supabase Postgres · Clerk · Razorpay

Applied AI

Generative AI · Claude API · OpenAI API · Agentic workflows · Multi-agent (Fetch.ai uAgents) · Tool use · Prompt engineering · Structured outputs (Zod) · Guardrails · Human-in-the-loop · LLM evals · LLM observability

Infra & Testing

Vercel · Netlify · GitHub Actions · Linux · Apache JMeter (load, race & failover tests) · Vitest

I build

Web apps · APIs & backends · Dashboards & SaaS · Auth & payments · AI features · Browser extensions

principles

How I build.

  1. 01

    AI drafts, humans confirm.

  2. 02

    State machines over vibes.

  3. 03

    Money as integer paise, never floats.

  4. 04

    A failed check is left out, never guessed.

open to work

What I’m looking for.

Roles

  • Applied AI engineer
  • Full-stack engineer
  • Software engineer (SDE-1)

Teams I’d fit

Commerce & merchant tools · WhatsApp & messaging · AI features in real products · Payments & checkout

Type
Full-time
Level
Entry level · B.Tech CSE, 2026
Location
Open to relocate
Start
Immediately

Email is the fastest way to reach me.

Email me Résumé LinkedIn