9 phases · 26 weeks · 180 days · ◪ weekly review · ⚒ project day
The curriculum
View the dependency map →Sequenced so nothing is assumed before it's taught. Every 7th day is a review checkpoint with spaced repetition and a shipped mini-project; the capstone thread starts on Day 119 and runs to Demo Day. Full design rationale lives in the repo's docs/curriculum-outline.md.
Phase 1
Code Foundations
days 1–21 · weeks 1–3Python from zero to fluent, plus the software-engineering habits — git, shell, testing, packaging, clean code — that everything else stands on.
D1Your Machine, Your MapPacking for the expeditionD2Variables, Types & Control FlowLabeled boxes and forks in the roadD3Functions, Scope & ModulesRecipes and kitchensD4CollectionsBookshelves and dictionariesD5Strings, Files & ErrorsPaper trails and safety netsD6The Terminal & LinuxThe workshopD7Week 1 Checkpoint: CLI Task Tracker ◪First checkpointD8Git — Your Time MachineSave points for your workD9Object-Oriented Python IBlueprints and housesD10Object-Oriented Python IIFamily trees vs LegoD11Iterators & GeneratorsConveyor beltsD12Closures, Decorators & Functional StyleGift-wrapping functionsD13Regex & Text ProcessingFind-and-replace with superpowersD14Week 2 Checkpoint: Log Analyzer ◪Second checkpointD15Clean Code & Type HintsA tidy workshopD16Logging, Config & CLI ErgonomicsThe flight recorderD17Packaging & EnvironmentsShipping the recipe boxD18Testing I — pytest FundamentalsSeatbelts and smoke alarmsD19Testing II & DebuggingThe crime-scene kitD20GitHub CollaborationWorking on the same houseD21Week 3 Checkpoint: Ship a Tested Package ◪Your first shipped tool
Phase 2
CS Core
days 22–49 · weeks 4–7Data structures, algorithms, SQL, operating systems, networking, concurrency, security, and system design — the invisible machinery under every production system.
D22Big O & ComplexityTwo chefs at a weddingD23Arrays & HashingNumbered shelves & straight-to-the-pageD24Two Pointers & Sliding WindowClosing pincers & the moving spotlightD25Stacks & QueuesTray piles and lunch linesD26Linked ListsTreasure huntsD27Recursion & Divide/ConquerRussian dollsD28Week 4 Checkpoint: Pattern Drill ◪The toolkit testD29Binary Trees & BSTsOrg charts with a sorting ruleD30Heaps, Priority Queues & TriesThe triage nurse & the autocomplete treeD31Graphs, BFS & DFSFriendship maps, ripples & corridorsD32SortingLining up the kidsD33Binary Search & VariantsHalving the phone bookD34Dynamic Programming IntroRemembering solved puzzlesD35Week 5 Checkpoint: Interview Drill I ◪The whiteboard warm-upD36SQL I — Tables & QueriesSpreadsheets with a contractD37SQL II — Joins & ModelingMatching guest listsD38Database InternalsThe index at the back of the bookD39Operating Systems EssentialsThe hotel managerD40Concurrency & Async PythonOne chef, many potsD41HTTP & Build Your First APIOrdering at the counterD42Week 6 Checkpoint: API + DB Mini-Service ◪The first backendD43Networking — Packets to HTTPSThe postal systemD44Security, AuthN & AuthZLocks, badges and guest listsD45Web Service ArchitectureThe restaurant's back of houseD46Distributed Systems FundamentalsMany kitchens, one restaurantD47Caching & QueuesThe pantry and the ticket railD48System Design MethodThe architect's rehearsalD49Week 7 Checkpoint: Design & Build a URL Shortener ◪First system design
Phase 3
Math for ML
days 50–63 · weeks 8–9Vectors, matrices, gradients, probability, statistics and entropy — exactly the math modern ML runs on, taught visually and by building.
D50Vectors & Dot ProductsArrows and shadowsD51Matrices & TransformationsMachines that move spaceD52Rank, Eigenvectors & SVD IntuitionThe grain of the woodD53Derivatives & GradientsThe sensitivity dialD54Chain Rule & OptimizationGears in a chainD55Gradient Descent LabRolling downhill in fogD56Week 8 Checkpoint: Linear Regression by Hand ◪The math becomes codeD57Probability FundamentalsWeather forecastsD58Random Variables & DistributionsThe shape of chanceD59Bayes' RuleUpdating your beliefsD60Statistics I — Sampling & ConfidenceTasting the soup, not drinking the potD61Statistics II — Hypothesis Tests & A/BThe courtroom standardD62Entropy, Cross-Entropy & KLSurprise as a currencyD63Week 9 Checkpoint: Math Assessment ◪The toolkit inspection
Phase 4
Data & Machine Learning
days 64–84 · weeks 10–12From messy CSV to validated model: NumPy, pandas, EDA, feature engineering, supervised and unsupervised learning, metrics, and honest error analysis.
D64NumPy in AngerPower tools for numbersD65pandas I — DataFramesThe spreadsheet that scriptsD66pandas II — WranglingThe data kitchenD67Exploratory Data AnalysisInterviewing your dataD68Cleaning & ValidationMise en placeD69Feature EngineeringCutting ingredients so the pan can cook themD70Week 10 Checkpoint: EDA Report ◪The data interview writeupD71ML Framing & Linear RegressionDrawing the best lineD72Logistic Regression & LossesConfidence, not just answersD73Decision TreesTwenty questionsD74Ensembles — Forests & BoostingAsking a crowdD75Evaluation MetricsGrading fairlyD76Validation, Bias/Variance & RegularizationPractice tests vs the real examD77Week 11 Checkpoint: Tabular Mini-Competition ◪First leaderboardD78ClusteringSorting a garage sale, unlabeledD79PCA & DimensionalityThe best camera angleD80Error AnalysisThe doctor reads the chartD81Experiment Tracking & ReproducibilityThe lab notebookD82ML Code Structure & PipelinesFrom notebook to factoryD83Phase Project: Churn Prediction End-to-End ⚒The whole assembly lineD84Week 12 Checkpoint + Interview Drill II ◪The ML viva
Phase 5
Deep Learning & Transformers
days 85–105 · weeks 13–15Neural networks from scratch, PyTorch, embeddings, attention, tokenization, and a tiny GPT you train yourself — the internals behind the APIs.
D85Neurons & Forward PassLayers of dialsD86Backpropagation from ScratchBlame flows backwardsD87PyTorch — Tensors & AutogradAutograd does your calculusD88Training Loops & DataThe practice scheduleD89Training DynamicsTuning the ovenD90MNIST Lab ⚒Hello, deep learningD91Week 13 Checkpoint — The First Neural Check ◪The first neural checkD92Embeddings — Meaning as GeometryThe map of meaningD93word2vec LabWords known by their companyD94AttentionEveryone looks at everyoneD95The TransformerThe assembly line of attentionD96TokenizationThe LLM's alphabetD97Tiny GPT Lab I — Build It ⚒Your own GPT, pocket-sizedD98Week 14 Checkpoint: Attention, Locked In ◪Closing the toolbox lidD99Tiny GPT Lab II — Train & Sample ⚒Watching it learn to spellD100How LLMs Are TrainedRaising a polymathD101Scaling Laws, Capabilities & LimitsBigger brains, sharper edgesD102Decoding & SamplingThe dice behind the wordsD103The Model Landscape & Local InferenceThe tool wallD104Context Windows & Hallucination Deep-DiveThe polymath's notepadD105Week 15 Checkpoint: Phase 5 Assessment ◪Internals exam
Phase 6
AI Engineering
days 106–133 · weeks 16–19The 2026 job core: LLM APIs, prompting, structured outputs, tool use, RAG, vector search, agents, MCP, fine-tuning, multimodal, and AI security.
D106LLM APIs I — The RequestRenting the polymathD107LLM APIs II — Streaming, Retries & CostThe meter is runningD108Prompt Engineering I — The Briefing MemoWriting the briefing memoD109Prompt Engineering II — Memos That Survive ContactMemos that survive contactD110Structured Outputs — Forms, Not EssaysForms, not essaysD111Tool Use & Function CallingGiving the polymath handsD112Week 16 Checkpoint: The Prompt Lab ◪The briefing-memo examD113RAG I — ArchitectureThe open-book examD114Chunking StrategiesTearing the book into useful pagesD115Embeddings & Vector DatabasesThe library with a meaning-based indexD116Hybrid SearchSmell plus card catalogD117Reranking & Query TransformsThe librarian double-checks the pileD118Advanced RAG PatternsBeyond one shelfD119Week 17 Checkpoint: Capstone Kickoff — Docs-QA v0 ◪The open-book exam, for realD120Agents I — The LoopAn intern with a to-do listD121Agents II — Planning & DecompositionBreaking the mission into missionsD122Agents III — Memory & ContextThe intern's notebookD123Multi-Agent & Workflow PatternsA small firm, not one internD124Model Context ProtocolThe universal adapterD125Agent ReliabilityTrust, but verifyD126Week 18 Checkpoint: Support-Triage Agent ◪The intern's first shiftD127Fine-Tuning I — When & DataTeaching the polymath your house styleD128Fine-Tuning II — LoRA LabSticky notes on the giant brainD129Distillation, Quantization & Model SelectionThe right size of brainD130Multimodal I — Vision & DocumentsEyes for the polymathD131Multimodal II — Audio & VoiceEars and a voiceD132Guardrails, Injection & AI SecurityThe con artist and the bank tellerD133Week 19 Checkpoint: Red-Team Your RAG ◪The heist rehearsal
Phase 7
Evaluation & Observability
days 134–147 · weeks 20–21The most underrated hiring signal: golden sets, LLM-as-judge, RAG metrics, regression gates, tracing, red teaming, and continuous evaluation.
D134Eval Mindset & Golden SetsThe exam written before the student existsD135Graders — Code, Rubric & LLM-as-JudgeWho grades the grader?D136RAG EvaluationGrading the open-book examD137Agent & Task EvalsDid the intern actually finish the job?D138Human EvaluationThe taste testD139Statistics for EvalsError bars or it didn't happenD140Week 20 Checkpoint: Capstone Eval Harness ◪The exam is now automatedD141Regression Gates & CI for AIThe tripwireD142Tracing LLM ApplicationsFlight recordersD143Logging, Feedback & the Data FlywheelEvery flight teaches the fleetD144Red-Teaming LabFire drillsD145Drift & Continuous Eval in ProdThe slow leakD146Quality & Cost DashboardsThe cockpitD147Observability Complete ◪Instruments all green
Phase 8
Production & MLOps
days 148–161 · weeks 22–23Docker, cloud, CI/CD, serving and inference optimization, monitoring, reliability, incident response — shipping AI that stays up and earns its cost.
D148Docker FundamentalsShipping the whole kitchenD149Compose, Registries & Image HygieneThe fleet manifestD150Cloud FundamentalsRenting racks by the minuteD151Deploy Lab — Container to Cloud URL ⚒Opening the doorsD152CI/CD with GitHub ActionsThe robot release managerD153Infrastructure as Code & EnvironmentsBlueprints for the building itselfD154Week 22 Checkpoint: Staging Pipeline ◪The assembly line runsD155Serving & Inference OptimizationThe drive-through windowD156Caching, Batching & Cost EngineeringThe pantry, again — at scaleD157Monitoring & SLOsSmoke alarms for softwareD158Reliability & Incident ResponseWhen the kitchen catches fireD159Production Security & Compliance BasicsLocking up at nightD160AI System DesignThe architect's examD161Week 23 Checkpoint: Production Cutover ◪Opening night
Phase 9
FDE & Capstone
days 162–180 · weeks 24–26Forward-deployed craft: discovery, specs, prototypes, enterprise integration, stakeholder communication — then ship and present your capstone.
D162The FDE RoleThe embedded field engineerD163Customer Discovery & the Mom TestQuestions that can't lie to youD164Ambiguity → RequirementsFog into blueprintsD165Proposals & Architecture DocsDrawing the blueprint togetherD166Rapid PrototypingThe movie trailer, not the movieD167Demo CraftThe show must go onD168Week 24 Checkpoint: FDE Simulation I ◪The client roomD169Enterprise IntegrationPlumbing into an old buildingD170Customer EnvironmentsCooking in someone else's kitchenD171Stakeholders & Trade-off NavigationTranslating between two languagesD172Production Debugging with CustomersThe field mechanicD173ROI, Pricing & Cost AnalysisIs the robot worth it?D174FDE Simulation II — Full Cycle ⚒The engagementD175Week 25 Checkpoint: Debrief & Gap Closure ◪The after-action reviewD176Capstone Hardening ⚒Punch-list week beginsD177Capstone Quality Gates ⚒The inspector's visitD178Capstone Ship & Document ⚒Cutting the ribbonD179Interview GymTraining camp, final weekD180Demo Day ◪Graduation