// ls ai-engineering/
AI Engineering
A systematic curriculum for engineers transitioning into AI engineering, built on distributed systems intuition.
Foundations 2 / 2
Retrieval & RAG 7 / 7
- 03 Vector Databases & ANN Indexes
- 04 Chunking Strategies for Retrieval
- 05 Hybrid Search: BM25 Meets Dense Vectors
- 06 Reranking: Cross-Encoders and Cascades
- 07 Query Transformations: Rewriting, HyDE, Multi-Query
- 08 RAG Evaluation: Recall, Faithfulness, Answer Quality
- 09 Context Engineering: JIT vs AOT Context Loading
Generation Control 5 / 5
Agents 8 / 8
- 15 The Agent Loop: ReAct and Its Descendants
- 16 Planning Agents vs Reactive Agents
- 17 Multi-Agent Orchestration
- 18 Tool Selection at Scale: MCP and Dynamic Tool Routing
- 19 Computer Use and Browser Agents
- 20 Long-Horizon Task Reliability
- 21 Anatomy of an Agent Harness
- 22 Conversation Compaction: Keeping Long Sessions Alive
Memory 21 / 21
- 23 The Memory Stack: A Map of AI Memory
- 24 The Cognitive Taxonomy: Semantic, Episodic, Procedural
- 25 Short-Term Memory: Managing the Conversation Buffer
- 26 Working Memory: Scratchpads, Blackboards, and Agent Notebooks
- 27 Long-Term Memory: Vector-Backed Episodic Storage
- 28 Knowledge Graphs as Structured Memory
- 29 Hierarchical Memory: Working / Episodic / Semantic Tiers
- 30 Memory Write Policies: What's Worth Remembering
- 31 Episode Segmentation and Salience Scoring
- 32 Reflection: From Experiences to Beliefs
- 33 Summarization and Context Compression
- 34 Sleep-Time Compute and Memory Consolidation
- 35 Memory Retrieval Policies: Recency, Relevance, Importance
- 36 Temporal Reasoning and Memory Provenance
- 37 Memory Conflict, Forgetting, and Embedding Drift
- 38 Procedural Memory and Skill Caching
- 39 Cross-Session Identity and Personalization
- 40 Multi-Agent Shared Memory
- 41 Memory Privacy, Isolation, and Multi-Tenancy
- 42 Memory Evaluation: Benchmarks and Custom Evals
- 43 Production Memory Frameworks: MemGPT/Letta, mem0, Zep, Graphiti