AI Engineer

Ramesh Kumar Voodi

RAG applications, agents, and developer tooling.

I work across model integration, agents, retrieval, evaluation, product engineering, and deployment, with seven public projects across AI systems, developer tools, and full-stack apps.

7
open-source projects shipped
Current work
AI systems, full-stack products, mobile apps, and developer tools

Selected work

Seven projects across AI systems, developer tools, web, mobile, and Web3.

01

Desktop AI workspace

Coworker

A cross-platform workspace that puts chat, a terminal, file navigation, and ripgrep search in one application.

Rust file operations through napi-rs, with a React 19 UI.

  • Rust
  • TypeScript
  • napi-rs
02

Agent runtime

Agentflow

A LangGraph runtime with retrieval, MCP tools, server-sent event streaming, and an evaluation harness.

Chroma RAG, YAML evaluation cases, and Docker CI live in the same runtime.

  • Python
  • LangGraph
  • RAG
  • MCP
03

Language model

smol-lm

A character-level GPT built from scratch in PyTorch, trained on Shakespeare with causal attention and transformer blocks.

4.97 val perplexity vs 11.96 bigram baseline; FastAPI serving with eval scripts.

  • Python
  • PyTorch
  • Transformer
04

Developer tool

Bloomery

A Rust CLI that converts traces from Claude Code, Codex, Cursor, and Pi into fine-tuning JSONL.

One binary, streaming parser, PII cleaning, and no Python runtime.

  • Rust
  • CLI
  • AI tooling
05

Workflow engine

Velum

Durable workflow orchestration with a gRPC API, event-sourced execution, and persistent state.

Built in Go with PostgreSQL-backed workflow state.

  • Go
  • gRPC
  • PostgreSQL
06

Mobile product

Polyswipe

A swipe-first mobile interface for exploring crypto prediction markets and live market data.

60fps Reanimated gestures with responsive live charts.

  • React Native
  • Expo
  • Mobile
07

Web3 wallet

Shade Wallet

A non-custodial Solana wallet with burner wallets, passkey login, and local key management.

Seed material is encrypted entirely on the client.

  • Solana
  • TypeScript
  • Web3

About

I build LLM agents, RAG systems, and developer tools, and I have built and shipped client software independently since 2024. My work spans architecture, implementation, deployment, and the debugging that keeps a system healthy.

I'm open to AI engineering and software engineering roles where I can own a product end to end, remote or with relocation.

Based
Visakhapatnam, India
Experience
AI systems, full-stack, mobile, and desktop apps
Work
Remote or relocation
Response
Usually within one day

Working stack

Technologies used across recent projects.

AI systems

  • LangGraph
  • LangChain
  • RAG
  • MCP
  • OpenAI
  • pgvector
  • Chroma
  • LangSmith

Languages

  • Python
  • TypeScript
  • JavaScript
  • Go
  • Rust
  • Swift

Full-stack and mobile

  • Next.js
  • React
  • React Native
  • Tailwind CSS
  • Node.js
  • FastAPI
  • PostgreSQL
  • MongoDB
  • gRPC

Infrastructure & APIs

  • Docker
  • Git / CI-CD
  • Vercel
  • Twilio
  • Stripe
  • Solana

Have a role or product that fits?

Let's talk.

rameshvoodi24@gmail.com