{
 "schema_version": 1,
 "generated": "2026-10-20",
 "repo": "patchy631/time-to-first-token",
 "name": "time-to-first-token",
 "owner": "patchy631",
 "slug": "patchy631-time-to-first-token",
 "github": "https://github.com/patchy631/time-to-first-token",
 "what": "A 10-week self-study roadmap for running LLM inference serving in production, built around one OpenAI-compatible service that gets deployed, instrumented, load tested, optimized, and benchmarked.",
 "why": "",
 "language": "HTML",
 "requirements": [
  "Python and command-line comfort",
  "One rented GPU with about 24GB of memory for a 7-8B model, billed by the half hour",
  "Docker for the Prometheus and Grafana stack",
  "An H100 for the 1000-concurrent load test in week 5"
 ],
 "audience_note": "Engineers who know Python, transformers at the architecture level, and the command line, and want hands-on LLM serving experience on rented GPUs.",
 "risk": "none",
 "section": "earlier",
 "weekly_rank": null,
 "category": "Software and DevOps",
 "url": "https://dedupedai.com/use-cases/patchy631-time-to-first-token/",
 "use_case_count": 4,
 "use_cases": [
  {
   "id": "patchy631-time-to-first-token--serve-openai-compatible-llm-on-rented-gpu",
   "title": "Serve a 7-8B model behind an OpenAI-compatible endpoint",
   "best": true,
   "url": "https://dedupedai.com/use-cases/patchy631-time-to-first-token/serve-openai-compatible-llm-on-rented-gpu/",
   "json": "https://dedupedai.com/use-cases/u/patchy631-time-to-first-token/serve-openai-compatible-llm-on-rented-gpu.json"
  },
  {
   "id": "patchy631-time-to-first-token--chart-ttft-and-queue-depth-grafana",
   "title": "Chart TTFT and queue depth on a live LLM service",
   "best": false,
   "url": "https://dedupedai.com/use-cases/patchy631-time-to-first-token/chart-ttft-and-queue-depth-grafana/",
   "json": "https://dedupedai.com/use-cases/u/patchy631-time-to-first-token/chart-ttft-and-queue-depth-grafana.json"
  },
  {
   "id": "patchy631-time-to-first-token--compare-vllm-sglang-shared-prefix-workload",
   "title": "Compare vLLM and SGLang on a shared-prefix workload",
   "best": false,
   "url": "https://dedupedai.com/use-cases/patchy631-time-to-first-token/compare-vllm-sglang-shared-prefix-workload/",
   "json": "https://dedupedai.com/use-cases/u/patchy631-time-to-first-token/compare-vllm-sglang-shared-prefix-workload.json"
  },
  {
   "id": "patchy631-time-to-first-token--load-test-llm-endpoint-past-1000-concurrent",
   "title": "Load test an LLM endpoint past 1000 concurrent requests",
   "best": false,
   "url": "https://dedupedai.com/use-cases/patchy631-time-to-first-token/load-test-llm-endpoint-past-1000-concurrent/",
   "json": "https://dedupedai.com/use-cases/u/patchy631-time-to-first-token/load-test-llm-endpoint-past-1000-concurrent.json"
  }
 ],
 "disclaimer": "Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk."
}