The FDE concept map
Follow an idea from one topic to the next. Explore how retrieval connects to evaluation, how agents connect to security, and where your next lesson fits. Select a concept to see its relationships or open the whole library.
191 concepts10 topics250 connections across topics
FOLLOW AN IDEA
See what connects.
Select a card to follow its connections. Lines show related concepts, not a required study order.
Explore a well-connected concept
Where the map is dense
The most-connected concepts are the ones the rest of the library keeps reaching for, which makes them useful places to build understanding of the library: Retrieval-Augmented Generation (RAG) (25), Golden Datasets and Eval Sets (24), Guardrails (21), Agentic Evals: Grading the Trajectory, Not the Answer (17), and Observability for AI Systems (17). If you are deciding where to spend a week, start with a hub rather than a leaf.
Every concept, by track
Foundations of LLMs & GenAI35
- Embeddings & Vector Representations13
- KV Cache13
- Model Routing and Cascades11
- The Transformer, Intuitively10
- Fine-tuning vs RAG vs Prompting9
- RLHF (Alignment)9
- Reward Models9
- The Context Window8
- Why LLMs Hallucinate8
- Structured Output and Schema Validation8
- System One Decision Models (Jev)8
- Prompt Engineering7
- Tokenization & Tokens6
- Temperature, Top-p and Sampling6
- Inference-Time Compute6
- Multimodal Models and VLMs6
- Dirty Looper Models: Looped Reasoning and Chain-of-Thought Monitorability6
- Web-Grounded Decisions (RAG for Decision Models)6
- Direct Preference Optimization (DPO)5
- Prompt Caching and Semantic Caching5
- Model Selection for Enterprise Deployments5
- Agent Training Loss Masks5
- On-Policy Agent Distillation5
- Attention and Self-Attention4
- Constrained Decoding4
- Chain-of-Thought Prompting4
- Constitutional AI and RLAIF4
- Policy Optimization: PPO and GRPO4
- RoPE and Positional Encodings3
- LoRA and Parameter-Efficient Fine-tuning3
- Mixture of Experts (MoE)3
- Scaling Laws3
- Speech and Voice AI3
- Autoregressive Decoding3
- Diffusion Models2
Retrieval & Agents37
- Retrieval-Augmented Generation (RAG)25
- Guardrails21
- Agentic Evals: Grading the Trajectory, Not the Answer17
- MCP (Model Context Protocol)16
- Tool / Function Calling13
- Bounded Autonomy and Human-in-the-Loop12
- Vector Databases11
- The ReAct Loop (Reason, Act, Observe)11
- Chunking Strategies10
- AI Agents and Tool Use10
- Agent vs Workflow vs a Single Call10
- Agent-to-Agent Interoperability (A2A)10
- Agent Observability and Telemetry (OpenTelemetry GenAI)9
- Hybrid Search (Lexical + Vector)8
- Reranking and Two-Stage Retrieval8
- Agent Memory8
- Permission-Aware RAG8
- Agent Gateways (AI, LLM and MCP Gateways)8
- Agent Gyms: Environments, Verification and Training8
- Agent Registries: Discovery, Inventory and the Kill Switch7
- Approximate Nearest Neighbor (ANN)6
- Agent Frameworks and How Agents Fail6
- Embedding Versions and Drift6
- Index Freshness and Staleness Windows6
- AG-UI: The Agent-User Interaction Protocol6
- Text-to-SQL6
- Semantic Layer for AI Agents (Agent Context Layer)6
- A2UI and Generative UI for Agents6
- Multi-Agent Orchestration5
- Context Window Management for FDE Agents5
- Context Failure Modes5
- GraphRAG and Contextual Retrieval5
- Query Rewriting, Expansion and HyDE5
- Late-Interaction Retrieval (ColBERT)5
- AP2: The Agent Payments Protocol5
- Document Parsing and Extraction5
- TF-IDF and BM253
Evaluation & ML Foundations27
- Golden Datasets and Eval Sets24
- LLM-as-a-Judge16
- A/B, Canary and Shadow Testing11
- Offline vs Online Evaluation11
- Precision, Recall and F110
- Gradient Descent & Learning Rate8
- Bias-Variance Tradeoff7
- Calibration and Uncertainty7
- Benchmarks and Their Limits7
- The Eval Loop: From Reading Traces to Watching Production7
- Choosing the Evaluator: Code Checks, LLM Judges and Human Review7
- Information Theory for ML: Entropy, Cross-Entropy, KL and Perplexity6
- Overfitting and Regularization6
- Evaluating RAG Systems6
- Error Analysis: Read the Traces Before Choosing a Metric6
- Faithfulness vs Answer Relevancy5
- Synthetic Data Generation4
- Catastrophic Forgetting3
- Loss Functions3
- Activation Functions3
- Neural Network Basics: Perceptron to MLP3
- Semi-Supervised and Self-Training3
- Computer Vision: Classification, Detection, Segmentation3
- Multi-Armed Bandits2
- Normalization: Batch vs Layer2
- Handling Imbalanced Data2
- Convex vs Non-Convex Optimization2
System Design for AI in Production20
- Observability for AI Systems17
- Latency Optimization16
- VPC and Air-Gapped Deployment13
- From Proof-of-Concept to Production12
- The Ontology (Semantic Layer)11
- AI Cost and Unit Economics10
- Circuit Breakers and Backpressure10
- Idempotency8
- Rate Limiting8
- Retries, Exponential Backoff and Jitter7
- SLOs, SLIs and Error Budgets7
- The Walking Skeleton (Thin Slice First)6
- REST API Design for Integrations6
- Containers and Kubernetes for Customer Deployments6
- Consistency, CAP and What Your Workflow Actually Needs6
- Deployment Models: SaaS, BYOC, On-Prem and Air-Gapped6
- The Enterprise AI Reference Architecture6
- Message Queues and Pub/Sub5
- Fallbacks and Provider Failover5
- Palantir's Platform: Foundry, AIP, Gotham and Apollo5
Data & SQL Engineering14
- Data Quality and Validation15
- Idempotent Data Pipelines12
- ETL vs ELT10
- Query Optimization and Execution Plans10
- Change Data Capture (CDC)8
- Lakehouse, Delta and the Medallion Architecture8
- Batch vs Streaming7
- Dimensional Modeling and Slowly Changing Dimensions6
- Orchestration: DAGs, Retries and Backfills6
- Spark Internals and Performance Tuning5
- SQL vs NoSQL: Choosing a Data Store5
- Deduplication and LSH4
- SQL Window Functions3
- Gaps and Islands3
AI Security, Privacy & Governance18
- RBAC, ABAC and Identity at Runtime for Agents15
- Audit Trails and Traceability14
- Multi-Tenancy and Data Isolation12
- OAuth 2.0, JWTs and Service Credentials9
- AI Incident Response9
- Prompt Injection and Defense8
- PII Handling and Redaction8
- AI Governance (SOC2, EU AI Act)8
- IAM and Least Privilege8
- Enterprise SSO: SAML and OIDC7
- Data Residency and Sovereignty7
- Secrets Management7
- Agent Sandboxing and Execution Isolation7
- Agentic Vulnerability Remediation7
- Agent Skill Supply-Chain Security6
- Mechanistic Interpretability4
- Differential Privacy3
- Federated Learning3
The Customer-Facing Craft12
- Deployment Decision Rights9
- Requirements Discovery8
- Pilot Acceptance Criteria8
- Scoping Ambiguous Problems7
- AI Benefits Realization7
- Explaining Trade-offs to Non-Engineers6
- FDE Practice Charter6
- Structured FDE Assessment6
- FDE Capacity Planning6
- FDE Practice Economics5
- Stakeholder Management4
- Recovering a Failing Live Demo3
The map is for orientation. If you would rather be told what to do in order, the start-here path sequences the same material by background and stage, and the courses walk it front to back.