Build skills in the right order.
Choose a course track or browse the full catalog. Every course combines visual explanations, interactive learning, coding practice, and knowledge checks.
Pick a course track. Follow it in order.
Each track contains only courses available on LDS and recommends the order to take them.
Understand the model, build the surrounding system, then learn to measure and operate it.
8 courses in sequenceFind the exact skill you need.
Courses are grouped by how learners use them, not by launch date. Search a topic or narrow the catalog by domain and access.
Foundations
Start from first principles in SQL, Python, statistics, or language models.
SQL Fundamentals
Master essential SQL skills. Learn to query databases, filter data, aggregate information, and combine tables with JOINs.
Python Fundamentals
Master Python essentials for data science with interactive animations, live coding, and hands-on practice in your browser.
Statistics Foundations
Build statistical intuition through interactive visualizations. Master mean, median, mode, spread, and data visualization.
LLM Foundations
How large language models actually work, from BPE tokens through RoPE attention, scaling laws, DPO/GRPO, KV caches, and mechanistic interpretability. Eight modules, eight interactive animations, beginner to advanced practitioner.
Analytics
Build the querying, data-wrangling, and experimentation skills used in decision teams.
SQL Mastery
Interactive animations show you exactly how joins, window functions, and complex queries work under the hood.
Pandas Fundamentals
Master pandas for data science, DataFrames, selection, groupby, merging, reshaping, time series, and the performance patterns that separate hobbyists from professionals.
SQL Window Functions
Master the highest-leverage advanced SQL skill. OVER(), ranking, LAG/LEAD, frame clauses, multi-dialect quirks, and the performance patterns that make analytical SQL fast.
A/B Testing & Experimentation
From hypothesis to production. Master the statistical, engineering, and interpretive skills data scientists use to run experiments that hold up, plus modern techniques (CUPED, sequential testing, bandits).
Statistics & ML
Move from uncertainty and probability to models you can train and evaluate.
Probability for Data Scientists
The language of uncertainty, from coin flips to the math that powers every ML model. 8 modules with interactive simulations and the Streamora running dataset.
Machine Learning Fundamentals (scikit-learn)
The supervised-learning loop the way working data scientists run it, regression, classification, regularization, cross-validation, pipelines. Every model trains live in your browser on the Lendly P2P-lending dataset, built on scikit-learn 1.3.
AI systems
Build, adapt, evaluate, and operate modern LLM applications end to end.
Building RAG Systems & Vector Search
Build production retrieval-augmented generation from first principles, BM25, dense embeddings, hybrid search + RRF, chunking, MMR, cross-encoder reranking, HyDE, ANN indexes (IVF/HNSW), and rigorous evaluation (Recall@k, MRR, nDCG). Every concept measured before/after on one realistic corpus.
Building AI Agents
Build AI agents from scratch, the agent loop, tool use & function calling, planning, reflection/self-correction, memory, multi-agent systems, evaluation, and guardrails. The model is a deterministic stand-in; every piece of agent machinery runs for real on the Helpwell support agent.
AI Evals: Test, Measure & Ship LLM Apps
Evaluate LLM, RAG, and agent systems for real, error analysis & failure taxonomies, deterministic & semantic scorers, LLM-as-judge alignment (Cohen’s kappa), confidence intervals & power, McNemar regression tests, and a CI gate. Model outputs and judge verdicts are baked; every scorer, statistic, and gate runs for real on the Helpwell support assistant.
Build a Tiny LLM: From Tokens to Text
Build a working language model from scratch: tokenizer (BPE), embeddings and position, attention, the transformer block, decoding, training, and interpretability. Every concept runs live in your browser on StoryByte, a real ~1M-parameter GPT trained on TinyStories; watch it write an original story one real token at a time.
Fine-Tuning LLMs: Make the Model Yours
The post-training sequel to Build a Tiny LLM. Take StoryByte, the real ~1.09M-parameter GPT, and make it follow requests: training data design, SFT with loss masking, LoRA/QLoRA, DPO preference tuning, knowledge distillation, and int8 quantization, ending with the playbook for real 7B+ models. Every number measured; every training run reproducible on a laptop CPU via the companion repo.
Context Engineering & AI Workflows
Build one Helpwell support workflow from a naive prompt to a measured system: context windows, prompt contracts, structured outputs, budgeted packing, grounding, injection defense, routing, human review, and regression gates.
Production LLM Systems
Turn one working LLM call into a service people can depend on. Follow a support ticket from request to release while you add service promises, traces, traffic controls, cost routing, failure protection, tool permissions, incident response, and rollback.