Target audience: Technical co-founders and startup CTOs who need to move fast from API prototypes to production-deployed AI features, making cost-effective architectural decisions under time pressure
Goal: Build a working LLM-powered product from API integration through production deployment, with a clear understanding of cost structures, vendor tradeoffs, and scaling strategies.
This pathway is optimized for speed. Start with the applied chapters (APIs, prompting, RAG) to get a prototype running, then read the production and strategy chapters before scaling. Skip the foundations and training chapters entirely; you can return to them later if your product requires custom model work.
Chapter Guide
- Skim Ch 07: The Modern LLM Landscape (vendor selection, model capabilities) vendor selection and model capabilities
- Skim Ch 09: Inference Optimization (cost and latency levers) cost and latency levers before you scale
- Focus Ch 10: Working with LLM APIs (your starting point: build a prototype) your starting point: build a prototype fast
- Focus Ch 11: Prompt Engineering (make your prototype reliable) make your prototype reliable and consistent
- Focus Ch 12: Hybrid ML+LLM Architectures (when to use ML vs. LLM) decide when ML, LLM, or both is right
- Focus Ch 19: Embeddings and Vector Databases add semantic search to your product
- Focus Ch 20: RAG (most startups need this) most startups need RAG for knowledge access
- Skim Ch 21: Conversational AI chat features your users will expect
- Skim Ch 22: AI Agents (if building agent-based products) assess whether your product needs agents
- Skim Ch 23: Tool Use and Protocols (MCP integration) MCP integration for tool connectivity
- Skim Ch 28: LLM Applications real-world application patterns and blueprints
- Focus Ch 29: Evaluation (measure before you scale) measure before you scale or raise
- Focus Ch 30: Observability and Monitoring catch errors and costs early
- Focus Ch 31: Production Engineering (deploy with confidence) deploy with confidence on day one
- Focus Ch 32: Safety and Compliance (legal exposure) legal exposure you must address before launch
- Focus Ch 33: Strategy, Product and ROI (pricing, unit economics, vendor selection) pricing, unit economics, and vendor strategy
- Skim Ch 34: Emerging Architectures upcoming models that may shift your technical bets
- Optional Ch 35: AI and Society regulatory landscape and societal trends
- Focus Ch 36: From Idea to Product Hypothesis framing the AI product hypothesis, feasibility scoring
- Focus Ch 37: Building and Steering AI Products observe-steer loops, prototype iteration, AI coding assistants
- Focus Ch 38: Shipping and Scaling AI Products launch economics, provider portability, monitoring
- Appendix V: Tooling Ecosystem – survey the broader tooling ecosystem
- Appendix U: Docker and Containers – containerize services for rapid deployment
- Appendix S: Inference Serving – deploy and serve models cost-effectively
What Comes Next
Return to the Reading Pathways overview to explore other pathways, or proceed to FM.4: How to Use This Book for a quick orientation on conventions and callout types, then start reading.