Full-Stack & AI Engineer · Bengaluru, India

Building software
from idea to production.

I'm Jayash. I build LLM systems and the products around them — a multi-provider RAG pipeline over a 500K-document archive, an MCP server that turns an enterprise platform into 26 reusable agent tools, and the React front-ends, FastAPI services, and deploys that put them in front of users.

Open to Full-Stack & AI Engineer roles

See case studies Resume Get in touch
  • Current role AI EngineerSyneca · Aug 2025 – present
  • Experience 1 yr full-time+ freelance since Dec 2023
  • Location Bengaluru, IndiaOpen to remote or relocation
  • Looking for Full-time rolesFull-stack or AI engineering
Contact findjayash@gmail.com +91 91360 07794 Resume
01

Selected Work

All roles

SYNECA · Syneca

Current role

A smart archival system for enterprise data, with an in-product AI agent (SIA) that answers questions against the archive.

Problem
Half a million archived documents were searchable but not answerable. Answers had to be traceable back to source material, and the business didn't want the product welded to a single LLM vendor.
What I built
A multi-model RAG pipeline with ChromaDB retrieval and source-traced citations, behind a provider abstraction that lets SIA switch between OpenAI, Claude, and Gemini — and swap vector-DB backends — at runtime with no code changes. Then an MCP server exposing SYNECA as 26 tools, so any external agent or LLM client can drive the platform.
Result
Agent response time dropped 5x. Model choice became a config decision instead of a migration, and SYNECA stopped being a closed product — it's now callable tooling for any agent that speaks MCP.
  • 500K documents indexed
  • 5× faster responses
  • 26 MCP tools
  • 3 LLM providers, hot-swappable
  • Python
  • FastAPI
  • ChromaDB
  • RAG
  • MCP
  • OpenAI
  • Claude
  • Gemini
Internal product — code not public

2WENT6EX · Freelance

Shipped

A non-contemporary fashion e-commerce storefront, built and launched as the sole developer.

Problem
A fashion label needed a storefront that matched its identity rather than a stock theme — and product browsing on the first pass was slow enough to lose shoppers mid-session.
What I built
The whole thing, end to end: scope, design, build, testing, and deployment. I replaced the default data-fetching path with a GraphQL Storefront API integration so product search and detail views pulled exactly the fields they needed, on a responsive build verified across devices.
Result
Product search and detail loading got 8x faster. The site went live at 2went6ex.com and stayed the label's storefront — delivered solo, from an ambiguous brief.
  • 8× faster product search
  • 1 developer, full lifecycle
  • GraphQL
  • Storefront API
  • JavaScript
  • Responsive UI
  • E-commerce
Visit the live site
02

Work Experience

More
  1. Aug 2025 — Present On-site

    AI Engineer · Syneca

    • Shipped a multi-model RAG pipeline over SYNECA's 500K-document enterprise archive — ChromaDB retrieval with source-traced citations — cutting agent response time 5x.
    • Engineered a multi-provider abstraction so the SIA agent switches between OpenAI, Claude, and Gemini at runtime, with vector-DB backends swappable the same way — no code changes.
    • Designed and deployed an MCP server exposing SYNECA as 26 tools, consumable by any external agent or LLM client.
  2. Jan 2025 — Jun 2025 Freelance

    Frontend Developer · ZOOMBUY

    • Built an SEO-optimized Shopify storefront with a responsive, performance-tuned component library.
    • Improved page load speed 30% with React.lazy code splitting, bringing the bounce rate down through faster loads and better UX.
  3. Dec 2023 — Jul 2024 Freelance

    Full-Stack Developer · 2WENT6EX

    • Launched 2went6ex.com, a non-contemporary fashion storefront, as sole developer — scope, design, build, testing, and deployment.
    • Integrated a GraphQL Storefront API that delivered 8x faster product search and detail loading, on a responsive build verified across devices.
03

Things I've Shipped

Read the case studies

Self-directed builds, each one shipped and open source — a database client, an offline-first PWA on a Rust/WebAssembly core, a local-first file manager with on-device LLM workflows, and an automation agent driven over MCP.

Dextr

Dextr

A cross-platform desktop client for SQL, NoSQL, object storage, and HTTP APIs — browse, edit, and query every data source side by side.
Lotus

Lotus

A local Windows automation agent controlled via Telegram — manage apps, files, system actions, and messaging with secure, real-time commands.
Notilus

Notilus

A Finder-style, local-first file manager built in Flutter, with a built-in Ollama chat panel, a Quick-Look-style preview viewer, and a workflow editor for prompt chains against your files.
deskdazzle

deskdazzle

An offline-first, all-in-one 'swiss army knife' PWA — markdown, data conversion, tasks, image tools and a draggable desktop, powered by a Rust→WebAssembly core with optional Firebase sync.

2WENT6EX

Non-contemporary fashion e-commerce site, built and shipped end to end.
04

Skills

More

Languages

  • JavaScript
  • TypeScript
  • Python
  • Rust
  • C++
  • C#
  • Dart
  • SQL

Frontend

  • React.js
  • Next.js
  • Tailwind CSS
  • Framer Motion
  • Vite

Backend

  • Node.js
  • FastAPI
  • Express.js
  • Serverless (Vercel, AWS Lambda)

AI / ML

  • PyTorch
  • Hugging Face
  • OpenAI API
  • Anthropic API
  • Gemini
  • Ollama
  • LangChain
  • RAG

AI Agents

  • MCP
  • LLM agents
  • Multi-agent orchestration
  • Prompt engineering
  • Function calling
  • Embeddings
  • Vector search

APIs

  • REST
  • GraphQL
  • WebSockets
  • Postman

Cloud & DevOps

  • AWS
  • GCP
  • Vertex AI
  • Docker
  • GitHub Actions
  • CI/CD
  • Terraform

Systems

  • Rust
  • WebAssembly
  • Flutter
  • PWA / offline-first
  • Cross-platform desktop

Databases

  • PostgreSQL
  • MongoDB
  • Redis
  • ChromaDB

Tools

  • Git
  • GitHub
  • VS Code
  • Figma
  • Jupyter
05

About

More

I build full-stack web applications and AI-driven systems, and I like owning a feature end to end — the data model, the API, the interface, and the deploy.

Day to day I work as an AI Engineer: multi-provider RAG pipelines over vector databases, Model Context Protocol (MCP) servers that turn enterprise systems into reusable agent tooling, and cross-platform desktop applications with a Rust/WebAssembly core.

The web side is not a side note — React, Next.js, Node.js, and FastAPI are how the AI work actually reaches users, and I've shipped production front-ends and cloud deployments alongside the model plumbing.

I'm a Computer Science graduate, and I care about writing clean, practical code that solves real problems. Ambiguous requirement to deployed release is the part of the job I enjoy most.

Hiring for AI or full-stack?

I'm open to Full-Stack & AI Engineer roles and selective freelance work. Send me the problem you're trying to solve — I'll tell you straight whether I'm the right person for it.

Start a conversation