Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I'm Suman Madipeddi, an AI/ML Engineer with 2 years of experience building production systems from concept to launch. I specialize in multi-agent systems, MCP servers, full-stack development, and computer vision. I'm seeking a founding role to leverage my deep technical expertise and drive strategic growth.…

Suman Madipeddi

AI Engineer, Full Stack Developer, Data Scientist, +2





I'm Suman Madipeddi, an AI/ML Engineer with 2 years of experience building production systems from concept to launch. I specialize in multi-agent systems, MCP servers, full-stack development, and computer vision. I'm seeking a founding role to leverage my deep technical expertise and drive strategic growth.…

Available to hire

I’m Suman Madipeddi, an AI/ML Engineer with 2 years of experience building production systems from concept to launch. I specialize in multi-agent systems, MCP servers, full-stack development, and computer vision.

I’m seeking a founding role to leverage my deep technical expertise and drive strategic growth.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Language

English

Fluent

Work Experience

AI/ML Software Engineer at Stealth AI startup

October 1, 2025 - Present

Architected and deployed a full-stack document automation system through multi-agent orchestration (planning, retrieval, tool-use). Achieved 91% extraction accuracy on unstructured documents with DeepSeek-OCR, reducing processing time by 85%. Built a reinforcement-learning training pipeline using SWE-Bench to optimize multi-step reasoning agents, improving task completion rate from 40% to 70% across 100+ test scenarios.

AI and LLMs Engineer at Minor Chores

August 1, 2024 - October 1, 2025

Designed and fine-tuned LLaMA-3.1-8B with LoRA (rank-16) and 4-bit quantization on 4K+ proprietary examples, reducing inference costs by 78% while maintaining 89% accuracy; deployed with vLLM achieving ~120ms P95 latency. Built a RAG chatbot with Vertex AI and Pinecone serving over 3,000 queries per month, and developed a location- and task-based recommendation system improving task assignment accuracy by 86%. Led the development of a cross-platform React app deployed on AWS ECS Fargate with real-time messaging and payments, shipping to 1,000+ users in 2 releases.