Building production agentic systems, RAG pipelines, and evaluation harnesses for agents and LLMs. Also working on LLM post-training (fine-tuning, reinforcement learning) with PyTorch, and inference optimization. Learning Rust for the systems layer. Looking for inference and research engineering roles.

What I work on

• Agentic systems: LangGraph, Pydantic AI, Google ADK, MCP, A2A Protocol
• RAG pipelines: embedding models, vector databases, retrieval quality, context engineering
• Evaluation: RAGAS, DeepEval, G-Eval and LLM-as-a-Judge harnesses for RAG and agents
• Post-training: fine-tuning and reinforcement learning with PyTorch
• Inference optimization: serving, KV cache, quantization, speculative decoding

  github: Akshayredekar07
  huggingface: akshayredekar45
  linkedin: akshayredekar07
  kaggle: akshayredekar
  x: akshayredekar07
  mail: akshayredekar04@gmail.com

Tech Stack

GitHub Stats