Systems Engineer — Distributed Systems & AI Infrastructurehttps://twilighttechie.github.io/https://twilighttechie.github.io/posts/gpu-hardware-field-guide/https://twilighttechie.github.io/posts/gpu-hardware-field-guide/A comprehensive, from-zero guide to GPU hardware buzzwords: HBM, memory bandwidth, FP8/FP4, NVLink, H100 vs H200 vs B200 vs GB200, NVL72 racks, AMD Instinct, TPUs, Groq, Cerebras — plus what to actually run in a homelab and why datacenter GPU infra is genuinely hard.Tue, 21 Jul 2026 00:00:00 GMThttps://twilighttechie.github.io/posts/production-inference-full-deck/https://twilighttechie.github.io/posts/production-inference-full-deck/The complete slide deck covering the production inference optimization strategy: architecture, benchmarks, routing, and cost analysis.Sat, 18 Jul 2026 00:10:00 GMThttps://twilighttechie.github.io/posts/inference-architecture-spec/https://twilighttechie.github.io/posts/inference-architecture-spec/Architectural specification for a production inference optimization strategy for the agentic inference cloud.Sat, 18 Jul 2026 00:05:00 GMThttps://twilighttechie.github.io/posts/inference-mini-benchmark/https://twilighttechie.github.io/posts/inference-mini-benchmark/Hands-on benchmark of quantization levels on a 1.5B model: decode throughput, TPOT, and memory footprint compared across FP16, Q8_0, and Q4_K_M.Sat, 18 Jul 2026 00:02:00 GMThttps://twilighttechie.github.io/posts/agentic-inference-presentation/https://twilighttechie.github.io/posts/agentic-inference-presentation/Slide deck on optimizing LLM inference for the agentic era: quantization, KV cache, batching, and bandwidth-bound decoding.Sat, 18 Jul 2026 00:00:00 GMThttps://twilighttechie.github.io/posts/vllm-cpu-inference-guide/https://twilighttechie.github.io/posts/vllm-cpu-inference-guide/A complete, beginner-friendly guide to vLLM: what inference actually is, how to build vLLM from source on an Apple Silicon Mac with no GPU, every command explained, the three errors I hit and fixed, the flags that matter, and real throughput numbers from my living room.Sat, 18 Jul 2026 00:00:00 GMThttps://twilighttechie.github.io/posts/opencode-semantic-router/https://twilighttechie.github.io/posts/opencode-semantic-router/Cursor picks the right model for you automatically. OpenCode doesn't - yet. Here's a complete, production-shaped guide to adding intelligent auto model selection to OpenCode (or any OpenAI-compatible agent) using vLLM Semantic Router and AgentGateway.Thu, 09 Jul 2026 00:00:00 GMThttps://twilighttechie.github.io/posts/agentgateway-observability/https://twilighttechie.github.io/posts/agentgateway-observability/I Almost Built a Grafana Stack—Then AgentGateway Shipped Everything I Needed.Thu, 02 Jul 2026 00:00:00 GMThttps://twilighttechie.github.io/posts/cassandra-compression/https://twilighttechie.github.io/posts/cassandra-compression/How Cassandra Compression Actually Works (Chunks, Offsets, and Reads)Wed, 01 Jul 2026 00:00:00 GMThttps://twilighttechie.github.io/posts/agentgateway-semantic/https://twilighttechie.github.io/posts/agentgateway-semantic/Giving AgentGateway a Semantic Brain with vLLM Semantic Router - Inside My HomelabMon, 22 Jun 2026 00:00:00 GMThttps://twilighttechie.github.io/posts/traced-agent/https://twilighttechie.github.io/posts/traced-agent/I Traced Personal Agent's Source Code. Inside Was Pi... And It Dreams at 3 AM.Mon, 15 Jun 2026 00:00:00 GMThttps://twilighttechie.github.io/posts/deciding-ai/https://twilighttechie.github.io/posts/deciding-ai/I Built an AI That Decides Which AI to Talk To — Running 24/7 From My Living RoomWed, 10 Jun 2026 00:00:00 GMT