Open to opportunities

shrey sharma

@shreysharma

Message

I build multi-agent LLM systems, fine-tuning pipelines, and evaluation harnesses.

India

Message

What I'm looking for

I'm seeking AI/ML Engineer or Research Engineer roles focused on applied LLM and agent research, where I can build multi-agent systems, fine-tuning pipelines, and rigorous evaluation harnesses.

I build multi-agent LLM systems, fine-tuning pipelines, and rigorous evaluation harnesses for applied AI research. My work spans planning loops, tool-calling layers, RAG memory, closed-loop verification, and Dockerized model deployment.

In KARMA, I designed a five-agent code-optimization pipeline that achieved an 8.99× speedup over the PyTorch eager baseline and 1.08× over torch.compile on a consumer 6GB GPU. I also built a verification harness that delivered 0% false-positive correctness across reported results and a ChromaDB memory layer that improved agent performance from 0.26× to 1.09× on an identical task.

I also architected Repliq, an autonomous workflow automation platform, and fine-tuned Llama 2 on GSM8K with QLoRA and PEFT. I'm seeking AI/ML Engineer or Research Engineer roles in applied LLM and agent research.

Experience

Work history, roles, and key accomplishments

AI/ML Engineer

Self Employed

Developed multi-agent LLM systems, fine-tuning pipelines, and evaluation harnesses. Designed agent architectures end to end, including planning loops, tool-calling layers, RAG memory, and closed-loop verification.

Python PyTorch Hugging Face Transformers QLoRA PEFT Quantization RAG Prompt Engineering Multi agent orchestration Tool calling MCP Model Evaluation Benchmarking Docker fastAPI AWS MLFlow DVC CI CD Git Linux

Education

Degrees, certifications, and relevant coursework

University of Lucknow

Bachelor of Technology, Computer Science (AI Specialization)

2022 -

Pursuing a Bachelor of Technology in Computer Science with an AI specialization, covering machine learning, deep learning, and natural language processing.