Skip to content

The Arcflow Roadmaps

AI Engineering roadmap

Build a working map of AI, with every term explained before it is used.

Download darkDownload light

Arcflow Plus is coming — review drills, research breakdowns, more AI. Get one email at launch.

One email at launch. That's it.

The Arcflow Roadmaps

AI Engineering

Build a working map of AI, with every term explained before it is used.

  1. 1

    AI Foundations

    Every AI term explained before it's used — starting from zero.

    22 available
    1. 01What Is AI?
    2. 02AI vs Machine Learning vs Deep Learning vs Generative AI
    3. 03What Is A Model?
    4. 04Training vs Inference
    5. soonHow Models Learn
    6. 05Data, Datasets, Examples, And Labels
    7. 06What Is A Neural Network?
    8. 07Parameters And Weights
    9. 08What Is A Large Language Model?
    10. soonWhy Models Follow Instructions
    11. 09Tokens And Tokenization
    12. 10Prompts, Context, And Completions
    13. soonContext Windows And Their Limits
    14. soonSystem Prompts
    15. soonTemperature And Sampling
    16. soonStructured Outputs
    17. 11Training Data vs Context vs Memory
    18. 12Hallucinations
    19. 13Embeddings In Plain English
    20. 14Vectors In Plain English
    21. 15Semantic Meaning And Similarity
    22. 16Cosine Similarity
    23. 17Retrieval In Plain English
    24. 18RAG In Plain English
    25. 19Fine-Tuning vs Prompting vs Retrieval
    26. 20Multimodal AI In Plain English
    27. 21What Is An AI Agent?
    28. 22What Are Evals?
  2. 2

    Building With AI

    How AI engineers actually build: prompts, agents, retrieval, and evals.

    17 available
    1. soonPrompt Engineering Patterns
    2. 01Tool Use And Function Calling
    3. 02Agent Loops
    4. soonAgent Memory
    5. soonPlanning And Reasoning Patterns
    6. soonMulti-Agent Systems
    7. soonHuman-In-The-Loop
    8. soonAgentic AI In Plain English
    9. soonHarness Engineering In Plain English
    10. 03Model Context Protocol
    11. 04Context Engineering
    12. soonDocument Ingestion And Chunking
    13. 05Vector Embeddings
    14. 06Semantic Space
    15. 07Vector Search
    16. 08Vector Databases
    17. 09Hybrid Search
    18. 10Re-ranking
    19. 11Query Expansion
    20. 12Parent-Child Retrieval
    21. 13Multi-Vector Retrieval
    22. 14Context Compression
    23. 15Knowledge Graph RAG
    24. 16Agentic RAG
    25. 17LLM Evaluation
    26. soonLLM-As-Judge And Eval Datasets
  3. 3

    AI System Studies

    Complete production AI systems, from first bottleneck to working architecture.

    8 available
    1. 01ChatGPT-Style LLM Inference System
    2. 02Perplexity-Style RAG Search System
    3. 03Cursor-Style AI Coding Assistant
    4. 04Vector Database Search System
    5. 05Document Q&A System
    6. 06AI Agent Tool-Use System
    7. 07Recommendation Embedding Pipeline
    8. 08LLM Evaluation Platform
  4. 4

    Under The Hood

    Why the machinery behaves the way it does — optional depth, read when you're curious.

    19 available
    1. 01Transformer Architecture
    2. 02Attention
    3. 03Multi-Head Attention
    4. 04Masked Attention
    5. 05Positional Embeddings
    6. 06KV Cache
    7. 07LLM Inference Serving
    8. 08Flash Attention
    9. 09Paged Attention
    10. 10Speculative Decoding
    11. 11Quantization
    12. 12Distillation
    13. 13Mixture Of Experts
    14. 14Indexing Techniques For Vector Search
    15. 15ANN Indexes
    16. 16HNSW Indexes
    17. 17Search Execution Flow
    18. 18Supervised vs Unsupervised vs Self-Supervised Learning
    19. 19Loss, Optimization, And Gradient Descent
  5. 5

    Production AI

    Guardrails, security, observability, and cost — what keeps AI systems alive in production.

    soon