Skip to content
Ashutosh
  • Work
  • Case study
  • Services
  • Contact
Resume

I build LLM apps that survive production

Generative AI Engineer in Bengaluru and San Francisco. Open to work.

Get in touch
  • 15+ products shipped
  • 600+ engineers mentored
  • 1,000+ DSA problems solved

Certified by

  • Anthropic
  • Kestra
  • SAP
Selected work

Things I built,
and what they had to survive.

Every one shipped. Rank 1 at TCS AI Friday, 119 unit tests on Kavach.

Kavach

The first hour, and the ninety days after

cybercrime-assistant.vercel.app
Kavach: The first hour, and the ninety days after

Nexus Cold Chain AI

Agentic cold chain safety, Rank 1 at TCS AI Friday

linkedin.com
Nexus Cold Chain AI: Agentic cold chain safety, Rank 1 at TCS AI Friday

VaaniVerse

Voice AI contact centre simulator

github.com
VaaniVerse: Voice AI contact centre simulator

PostGen

Multi-agent LinkedIn ghostwriter

postgen-ashusnapx.vercel.app
PostGen: Multi-agent LinkedIn ghostwriter

AI Book Explorer

Conversational discovery over a book corpus

github.com
AI Book Explorer: Conversational discovery over a book corpus

Know About Your Food

Multimodal nutrition analysis

know-your-food-ashusnapx.streamlit.app
Know About Your Food: Multimodal nutrition analysis
Gemini VisionPythonStreamlit

AI Coal Mines Laws

Domain assistant for mining regulation

dgms-gemini-ashusnapx.streamlit.app
AI Coal Mines Laws: Domain assistant for mining regulation
Gemini ProDomain RAGPython

GenAI Apps Generator

Scaffolding for AI powered apps

genai-assignment.vercel.app
GenAI Apps Generator: Scaffolding for AI powered apps
Gemini ProNext.jsClerk

Habit.AI

Productivity tracking that sticks

habit-ai-lake.vercel.app
Habit.AI: Productivity tracking that sticks
Next.jsAppwriteshadcn/ui

Airbnb Clone

Listings, auth and booking flow

airbnb-by-ashusnapx.vercel.app
Airbnb Clone: Listings, auth and booking flow
ReactSupabaseOAuth

YouTube Playlist Analytics

How long is that course, really?

yt-playlist-length-4nzq.onrender.com
YouTube Playlist Analytics: How long is that course, really?
PythonDjango

Secourse

Study planning for aspirants

github.com
Secourse: Study planning for aspirants
React NativeNativewindAppwrite
Case study

Kavach

Eight obligations on statutory clocks, driven from a spoken account in any of 23 languages.

See it live
The Kavach triage screen: a spoken fraud report classified against the NCRP category tree, with the extracted identifiers and the statutory deadlines listed beside it.

23 languages in, classified against the official NCRP category tree.

Identifiers stripped by regex before any model sees them.

Seven documents out: portal complaint, 1930 script, bank dispute letter, FIR application.

Built under a clock,
and it still shipped.

See the resume

Rank 1, then the regionals

TCS AI Friday, Season 2, 2026

We built an agentic system for vaccine cold chain safety.

LangChainMulti-agentMCP

Shipped Kavach

Build What Moves India, 2026

The obvious build was a nicer version of the government complaint form.

Next 16GPT-4oStrict JSON schema

First hackathon, prototype after

Vaani Research Labs, Hackathon and meetup, 2026

My first hackathon, hosted with AI Learn Circle.

Voice AITypeScriptVaani APIs
Experience

How I
got here.

Enterprise systems, where breaking production has a price.

  1. Sep 2026 to Now

    Computer Vision Engineer

    Tata Consultancy Services (TCS)

    • Moving from Core ABAP onto computer vision work, applying the same production discipline to a model layer that fails in ways a business rule never does.
    • Building and evaluating vision models, and the pipelines that keep their outputs checked before anything downstream consumes them.
    • Computer Vision
    • Python
    • Deep Learning

  2. Jun 2025 to Sep 2026

    Software Engineer (Core ABAP)

    Tata Consultancy Services (TCS)

    • Developing and maintaining enterprise-grade business applications for a European client (Bayer) using Core ABAP.
    • Performed system analysis, debugging, testing, and defect resolution within SAP-based production systems.
    • ABAP
    • SAP
    • Enterprise Applications

  3. Dec 2023 to Jan 2024

    Frontend Developer

    AI Caller.io

    • Implemented CSV upload functionality to trigger bulk calling workflows.
    • Migrated application from client-side rendering to server-side rendering for improved performance and SEO.
    • NextJS
    • ReactJS
    • Redux Toolkit

  4. Feb 2022 to Oct 2022

    Teaching Assistant

    Coding Ninjas

    • Solved more than 1000 questions related to C++, Data Structures, and Algorithms.
    • Mentored and assisted more than 600 students in resolving coding and logical queries.
    • C++
    • Data Structures
    • Algorithms
The stack

What I reach for,
and what it is for.

Model layer first. The rest exists to make it usable.

Generative AI

the model layer

  • RAG
  • LangGraph
  • LangChain
  • CopilotKit
  • Multi-agent workflows
  • Tool calling
  • Structured outputs
  • Claude API
  • Gemini
  • OpenAI
  • MCP
  • Embeddings
  • Mem0
  • Braintrust evals
  • Pydantic AI
  • Voice AI

Languages

day to day

  • Python
  • TypeScript
  • JavaScript
  • C++
  • SQL
  • ABAP

Product and frontend

where the model meets a user

  • Next.js 16
  • React 19
  • Server Components
  • Streaming responses
  • Tailwind
  • Framer Motion
  • Redux Toolkit
  • React Native

Backend and data

what holds it up

  • Node.js
  • Express
  • Streamlit
  • PostgreSQL
  • Prisma
  • MongoDB
  • Supabase
  • Redis
  • REST APIs

Foundations

the part that does not go stale

  • Data structures
  • Algorithms
  • System design
  • NumPy
  • Pandas
  • Git
  • SEO
RAGLangGraphLangChainClaudeGeminiOpenAIMCPVector searchAgentsVoice AILLM evalsPrompt engineeringNext.jsTypeScriptPythonReactPostgreSQL
RAGLangGraphLangChainClaudeGeminiOpenAIMCPVector searchAgentsVoice AILLM evalsPrompt engineeringNext.jsTypeScriptPythonReactPostgreSQL
Credentials

Certified,
and checkable.

Every one links to its issuer. Go and check.

  • Claude Certified Developer, Foundations, issued by Anthropic

    Anthropic

    Claude Certified Developer, Foundations

    Aug 2026 to Aug 2027

    For developers who can build, integrate and ship production applications and agents on Claude using the Claude API, Claude Code, custom tools and MCP servers.

    Verify
  • Claude Code in Action, issued by Anthropic

    Anthropic

    Claude Code in Action

    Mar 2026

    Agentic development with Claude Code: driving multi-step engineering work through tool use and the Model Context Protocol.

    Verifyuc66j6vuqecs
  • Introduction to Agent Skills, issued by Anthropic

    Anthropic

    Introduction to Agent Skills

    Mar 2026

    Building and packaging Agent Skills, the reusable capability units that extend what an agent can do.

    Verifypk474prkjtf8
  • Kestra

    Kestra

    Kestra Fundamentals

    Declarative orchestration of data and workflow pipelines.

    Issuer verifiedYKXTMLod
  • SAP Certified Associate, Security Administrator, issued by SAP

    SAP

    SAP Certified Associate, Security Administrator

    Aug 2024

    Core SAP authorization and security concepts across S/4HANA Public and Private Edition.

    Verify
Live

Read straight
from GitHub.

Pulled from the API every time this page loads.

…
Public repos
…
Commits this year
…
Years on GitHub
…
Followers

Counting commits…

Languages

Three ways to work together

Full time

The one I am looking for

  • Generative AI engineering
  • RAG and retrieval pipelines
  • Multi-agent workflows
  • Production LLM apps

Open

Open to Generative AI engineering roles

Contract

Scoped builds, end to end

  • Everything in full time
  • Discovery and scoping
  • Evals and guardrails
  • Handover and docs

Project

From brief to shipped

Mentoring

Teams and individuals

  • Everything in contract
  • Code and architecture review
  • LLM app design sessions
  • Interview and DSA coaching

600+

Engineers mentored so far

  • 15+ products shipped
  • 1,000+ DSA problems solved
  • 600+ engineers mentored
  • 5 certifications
“I build production LLM applications: retrieval pipelines, multi-agent workflows and the products they ship inside.”
Ashutosh Kumar, Generative AI EngineerAshutosh Kumar, Generative AI Engineer

Frequently
asked questions

What comes up before a first call.

Ashutosh Kumar+You

Still have questions?

One email, one straight answer.

Talk to me

Generative AI engineering roles. I want the part after the demo: retrieval that holds up on real documents, agents that fail safely, evals that catch a regression before a user does.

The next thing I ship
is one I have not built yet.

Open to Generative AI engineering roles. Send the hard problem, I will ship it.

Get in touch
Ashutosh Kumar

Production LLM applications: retrieval pipelines and multi-agent workflows.

ashu.kumarexam@gmail.com

Quick links

  • Work
  • Case study
  • Hackathons
  • Certified
  • Stack
  • GitHub

Elsewhere

  • Hashnode
  • Medium
  • dev.to
  • Resume

Bengaluru and San Francisco