AI Engineering in Production

Ship and operate: serving, quantization (AWQ/GGUF), vLLM/KV cache, eval-gated CI/CD, cost and drift monitoring, LLMOps.

1. 1Serving models: latency and throughput3 q
2. 2Quantization: running models cheaply3 q
3. 3The KV cache and the memory cost of serving3 q
4. 4Evals as a production gate3 q
5. 5Cost monitoring and budgets3 q
6. 6Drift and quality monitoring3 q
7. 7The LLMOps lifecycle: versioning and rollback3 q
8. 8Shipping: serving, gating, and monitoring as one system3 q