AI/ML Engineer · Building agentic platforms, KV-cache & inference systems
Ahmedabad, Gujarat, India | 📧 rajveer.rathod1301@gmail.com
I'm an AI/ML Engineer working on multi-tenant agentic platforms, internal ML infrastructure, and client-facing ML solutions. My technical interests span LLM fine-tuning, KV-cache optimization, and production systems engineering.
- ⚡ Maintaining VeloxQuant-MLX, a KV-cache quantization library for Apple Silicon with 39+ eviction methods
- 🔬 Former HEP-ML researcher at Physical Research Laboratory — hypergraph neural networks for jet classification
- 🌱 Active open-source contributor — 35+ ML repositories including PyTorch and Hugging Face Transformers
- 🎓 B.Tech in Computer Engineering, Birla Vishvakarma Mahavidyalaya (GTU), 2023
- 📜 Linux Foundation PyTorch Certified (LFS116)
- ✍️ Write about ML systems & engineering on Medium
|
KV-Cache Quantization for Apple Silicon A library implementing 39+ KV-cache eviction methods for efficient LLM inference on Apple Silicon, with versioned releases and PyPI analytics tracking. |
|
|
Python VS Code Extension Call graph visualization tool with 6,700+ PyPI downloads and 190+ VS Code installs. |
35+ ML repositories including PyTorch and Hugging Face Transformers, plus regular participation in open-source events like Open Source Summit India and Open Source Day. |
Languages
ML / AI
Backend / APIs
Data
Tools
