Original Academy
Legacy mirror · noindex · upstream phase 17
Infrastructure & Production
Ship AI to the real world.
Provenance: this phase outline and its lesson readings are preserved from ai-engineering-from-scratch by Rohit Ghumare under the MIT License. InterviewsVector does not claim authorship. These archive pages remain available for old links and progress, but are excluded from indexing.
Attributed readings
- 01Managed LLM Platforms — Bedrock, Azure OpenAI, Vertex AIThree hyperscalers, three distinct strategies. AWS Bedrock is a model marketplace — Claude, Llama, Titan, Stability, Cohere behind one API. Azure OpenAI is an exclusive OpenAI p…
- 02Inference Platform Economics — Fireworks, Together, Baseten, ModalThe 2026 inference market is no longer GPU time rental. It bifurcates into custom silicon (Groq, Cerebras, SambaNova), GPU platforms (Baseten, Together, Fireworks, Modal), and A…
- 03GPU Autoscaling on Kubernetes — Karpenter, KAI SchedulerThree layers, not one. Karpenter provisions nodes dynamically (under one minute, 40% faster than Cluster Autoscaler). KAI Scheduler handles gang scheduling, topology awareness, …
- 04vLLM Serving Internals — PagedAttention, Continuous Batching, Chunked PrefillvLLM's dominance in 2026 rests on three compounding defaults, not a single trick. PagedAttention is always on. Continuous batching injects new requests into the active batch bet…
- 05EAGLE-3 Speculative Decoding in ProductionSpeculative decoding pairs a fast draft model with the target model. The draft proposes K tokens; the target verifies in a single forward; accepted tokens are free. In 2026, EAG…
- 06SGLang and RadixAttention for Prefix-Heavy WorkloadsSGLang treats the KV cache as a first-class, reusable resource stored in a radix tree. Where vLLM schedules requests FCFS (first-come, first-served), SGLang's cache-aware schedu…
- 07TensorRT-LLM on Blackwell with FP8 and NVFP4TensorRT-LLM is NVIDIA-only but it wins on Blackwell. On GB200 NVL72 with Dynamo orchestration, SemiAnalysis InferenceX measured $0.012 per million tokens on a 120B model in Q1-…
- 08Inference Metrics — TTFT, TPOT, ITL, Goodput, P99Four metrics decide whether an inference deployment is working. TTFT is prefill plus queue plus network. TPOT (equivalently ITL) is the memory-bound decode cost per token. End-t…
- 09Production Quantization — AWQ, GPTQ, GGUF, FP8, NVFP4Quantization format is not a universal choice — it is a function of hardware, serving engine, and workload. GGUF Q4_K_M or Q5_K_M owns CPU and edge, delivered through llama.cpp …
- 10Cold Start Mitigation for Serverless LLMsA 20 GB model image takes 5-10 minutes (7B) to 20+ minutes (70B) to go from cold to serving. In a true serverless world, that is not a warm-up — it is an outage. Mitigations ope…
- 11Multi-Region LLM Serving and KV Cache LocalityRound-robin load balancing is actively harmful for cached LLM inference. A request that does not land on the node holding its prefix pays full prefill cost — roughly 800 ms at P…
- 12Edge Inference — ANE, Hexagon, WebGPU, JetsonThe core edge constraint is memory bandwidth, not compute. Mobile DRAM sits at 50-90 GB/s; datacenter HBM3 clears 2-3 TB/s — a 30-50x gap. Decode is memory-bound so the gap is d…
- 13LLM Observability Stack SelectionThe 2026 observability market splits into two categories. Development platforms (LangSmith, Langfuse, Comet Opik) bundle monitoring with evals, prompt management, session replay…
- 14Prompt Caching and Semantic Caching Economics**Pricing snapshot dated 2026-04.** Numeric claims below reflect vendor rate cards captured at this lesson's publication; verify against the linked docs before quoting them down…
- 15Batch APIs — the 50% Discount as Industry StandardEvery major provider ships an async batch API with a 50% discount and ~24-hour turnaround. OpenAI, Anthropic, Google, and most of the inference platforms (Fireworks batch tier, …
- 16Model Routing as a Cost-Reduction PrimitiveA dynamic broker evaluates every request (task type, token length, embedding similarity, confidence) and sends simple queries to a cheap model, escalating complex ones to a fron…
- 17Disaggregated Prefill/Decode — NVIDIA Dynamo and llm-dPrefill is compute-bound; decode is memory-bound. Running both on the same GPU wastes one resource. Disaggregation splits them onto separate pools and transfers KV cache between…
- 18vLLM Production Stack with LMCache KV OffloadingvLLM's production-stack is the reference Kubernetes deployment — router, engines, and observability wired together. LMCache is the KV-offloading layer that extracts KV cache out…
- 19AI Gateways — LiteLLM, Portkey, Kong, BifrostA gateway sits between your apps and model providers. Core features are provider routing, fallback, retries, rate limiting, secret references, observability, guardrails. Market …
- 20Shadow, Canary, and Progressive DeploymentLLM rollouts combine the hardest parts of software deployment: no unit tests, diffuse failure modes, delayed signals. The sequence is (1) shadow mode — duplicate prod requests t…
- 21A/B Testing LLM Features — GrowthBook and StatsigTraditional A/B testing was not built for non-deterministic LLMs. The critical distinction: evals answer "can the model do the job?" A/B tests answer "do users care?" Both are r…
- 22Load Testing LLM APIs — k6, LLMPerf, GenAI-PerfTraditional load testers were not designed for streaming responses, variable output lengths, token-level metrics, or GPU saturation. Two traps bite most teams. The GIL trap: Loc…
- 23SRE for AI — Multi-Agent Incident ResponseAI SRE uses LLMs grounded in infrastructure data (logs, runbooks, service topology) via RAG to automate investigation, documentation, and coordination phases. The 2026 architect…
- 24Chaos Engineering for LLM ProductionChaos engineering for LLMs is its own discipline in 2026. Prerequisites before running experiments in production: defined SLI/SLO, trace+metric+log observability, automated roll…
- 25Security — Secrets, PII Scrubbing, Audit LogsEliminate secret sprawl via centralized vaults (HashiCorp Vault, AWS Secrets Manager, Azure Key Vault). Never store credentials in config files, env files in VCS, spreadsheets. …
- 26Compliance — SOC 2, HIPAA, GDPR, EU AI Act, ISO 42001Multi-framework coverage is table stakes for 2026 enterprise deals. **EU AI Act**: in force since August 1, 2024. Most high-risk requirements enforce August 2, 2026. Fines up to…
- 27FinOps for LLMs — Unit Economics and Multi-Tenant AttributionTraditional FinOps breaks on LLM spend. Costs are token-transactions, not resource-uptime. Tags don't map — an API call is a transaction, not an asset. Engineering decisions (pr…
- 28Self-Hosted Serving Selection — llama.cpp, Ollama, TGI, vLLM, SGLangFour engines dominate self-hosted inference in 2026. Pick based on hardware, scale, and ecosystem. **llama.cpp** is fastest on CPU — widest model support, full control over quan…