Pinned Loading

1. llm-inference-performance llm-inference-performance Public

   LLM inference performance engineering: a benchmark harness (p50/p95 latency, tokens/sec, memory) with measured before/after for KV-cache, batching, and int8 quantization, plus tensor-parallel servi…

   Python 1

2. inference-platform inference-platform Public

   An autoscaling LLM inference platform on Kubernetes: KEDA scaling on queue-depth/KV-cache via a custom Go external scaler, validated on a DIY two-GPU vLLM cluster.

   Go 1

3. typescript-template typescript-template Public template

   Starter template for Next.js Typescript apps

   TypeScript

4. python-template python-template Public template

   uv + ruff + pyright + pytest, centralized reusable CI, grouped Dependabot with auto-merge

   Makefile