Supercharge Your Model Training
Efficient Deep Learning Systems course materials
From teacher to tiles — a from-scratch LLM distillation & serving engine: custom Triton/CUDA kernels, FSDP distillation, paged-KV continuous batching, speculative decoding, a Rust gateway, a JAX oracle, and interpretability tooling.
Run more RL experiments. Wait less for GPUs.
Multimodal RAG system with CLIP/text embeddings, rank-fusion reranking, FastAPI serving, and Recall@5 evaluation.
Designing IT and ML Applications using Systems Thinking Approach at IIT Bhilai (CS559)
A curated list of resources for ML Systems Engineering - hardware, compilers, distributed training, inference, and production operations.
Measure and audit intelligence gained per joule spent. Proof-of-Learning prototype.
LICITRA v1 evidence — superseded by licitra-mmr-evidence
LICITRA v1 — superseded by licitra-mmr-core
Structured notes on designing scalable and fault-tolerant ML systems, to refresh your knowledge and help you prepare for a system design interview. Covers system design, MLOps, and case studies.
Rust + cuTile research prototype for paged latent-cache LLM decode attention, validated on an RTX 4060.
One small, honest ML system per day, built end to end on Hopsworks.
A compact C++17/CUDA offline LLM inference engine with scheduling, paged KV cache, Hugging Face loading, and reproducible benchmarks.
Autonomous platform for open-vocabulary perception, target acquisition, edge inference, and reproducible real-world ML experiments.
End-to-end implementations of 80+ landmark ML research papers in Python, C++,PyTorch and R — designed with a production-first mindset for scalable ML systems.
A curated research atlas for Efficient ML, AI Infrastructure, and Physical AI
Open research infrastructure for measuring capability-efficiency tradeoffs in post-trained small open-weight language models.
To associate your repository with the ml-systems topic, visit your repo's landing page and select "manage topics."