Programming
AI for Developers
Coding Assistants, AI Features and Good Engineering
- Pages
- 85
- Chapters
- 72
- Level
- Intermediate
- Format
- E-book
About this book
AI changes software development in two ways: how we write code, and what our code can do. This book covers both.
The first part shows how to work productively with coding assistants and agents: giving context, reviewing generated code, testing, refactoring, debugging and keeping security and licensing in view. The second part covers building AI features: calling models through APIs, structured output, retrieval, tool use, streaming, evaluation, cost and latency management, and production concerns.
Examples are language-neutral where possible, with short Python and JavaScript snippets.
What you will learn
- Working effectively with coding assistants and agents
- Reviewing, testing and securing generated code
- Calling models: prompts, structured output and streaming
- Retrieval and tool use in applications
- Evaluating and monitoring AI features
- Managing cost, latency and reliability in production
Table of contents
- IntroductionIntroduction
- Chapter 1The Developer's Mental Model
- Chapter 2Tokens and Context Windows
- Chapter 3A Basic API Call
- Chapter 4Statelessness and Conversation History
- Chapter 5Temperature and Determinism
- Chapter 6Choosing a Model
- Chapter 7Hosted APIs Versus Self-Hosting
- Chapter 8Latency
- Chapter 9Errors and Retries
- Chapter 10An Abstraction Layer
- Chapter 11Prompts as Code
- Chapter 12System Prompt Structure
- Chapter 13Delimiting Untrusted Input
- Chapter 14Few-Shot Examples
- Chapter 15Structured Output
- Chapter 16Handling Invalid Output
- Chapter 17Tool Calling
- Chapter 18Designing Good Tools
- Chapter 19Validating Tool Calls
- Chapter 20Streaming
- Chapter 21Multimodal Inputs
- Chapter 22Prompt Caching
- Chapter 23Why Retrieval-Augmented Generation
- Chapter 24The RAG Pipeline
- Chapter 25Chunking
- Chapter 26Embeddings and Vector Search
- Chapter 27Hybrid Search
- Chapter 28Reranking
- Chapter 29Permissions in Retrieval
- Chapter 30Citations and Grounding
- Chapter 31Evaluating Retrieval
- Chapter 32Keeping the Index Fresh
- Chapter 33Query Rewriting
- Chapter 34When Not to Use RAG
- Chapter 35Workflows Versus Agents
- Chapter 36Common Workflow Patterns
- Chapter 37The Agent Loop
- Chapter 38Agent Tool Design
- Chapter 39Planning and Reflection
- Chapter 40Human-in-the-Loop
- Chapter 41Memory for Agents
- Chapter 42Multi-Agent Systems
- Chapter 43Sandboxing
- Chapter 44Interoperability Protocols
- Chapter 45Debugging Agents
- Chapter 46Agent Reliability
- Chapter 47Why Evaluation Matters
- Chapter 48Building an Evaluation Set
- Chapter 49Grading Methods
- Chapter 50Model-Graded Evaluation
- Chapter 51Regression Testing
- Chapter 52Unit Tests Around the Model
- Chapter 53Observability
- Chapter 54Production Monitoring
- Chapter 55User Feedback Loops
- Chapter 56A/B Testing
- Chapter 57Model Upgrades
- Chapter 58Documentation for AI Features
- Chapter 59Prompt Injection
- Chapter 60Data Leakage
- Chapter 61Output Handling
- Chapter 62Secrets and Keys
- Chapter 63Cost Control
- Chapter 64Caching Responses
- Chapter 65Batch Processing
- Chapter 66Deployment Patterns
- Chapter 67Worked Example: A Help-Centre Answer Feature
- Chapter 68Worked Example: Invoice Extraction Pipeline
- Chapter 69Coding Assistants
- Chapter 70Working Well With Coding Assistants
- Chapter 71Reviewing AI-Generated Code
- Chapter 72A Production Readiness Checklist
- GlossaryGlossary
Who this book is for
Software developers who want to use AI in their workflow and build AI-powered features well.
Details
| Format | PDF + web edition |
|---|---|
| Length | 85 pages (approx.) |
| Chapters | 72 |
| Level | Intermediate |
| Language | English |
| Topic | Programming |
| Author | Latent Library Editorial |
Keep reading