Full curriculum

BuildAfreeka_ — Everything we teach, in sixteen months.

This includes the full AI & Agents track, layered on top of the fundamentals — students don't skip the basics to get there. Every phase below breaks down into named modules with the actual topics covered.

MONTHS 1–6 — Core software fundamentals MONTHS 7–9 — Applied AI & LLM foundations MONTHS 10–12 — Agent building MONTHS 13–16 — Evaluation, production & Build Studio
MONTHS 1–6

Core software fundamentals

Everyone starts here, regardless of background — this is where aptitude turns into ability.

MODULE 1.1 — MONTHS 1–2

Programming Foundations

Learn to think and build in code, from first principles.

  • Variables, control flow, functions
  • Object-oriented and basic functional programming
  • Python as the primary teaching language
  • Computational thinking and problem decomposition
MODULE 1.2 — MONTHS 2–3

Data Structures & Algorithms

The toolkit every working developer reaches for, taught through practical problems, not just theory.

  • Arrays, lists, hash maps, trees, graphs
  • Sorting and searching
  • Time and space complexity (Big-O)
  • Technical interview-style problem drills
MODULE 1.3 — MONTHS 3–5

Web & API Development

Build real, working software — the backbone skill for every later phase.

  • HTTP, REST APIs, and client-server architecture
  • Relational databases and SQL
  • Backend development with FastAPI
  • Frontend fundamentals: HTML, CSS, JavaScript
MODULE 1.4 — THROUGHOUT

Version Control & Collaboration

How real engineering teams actually work together, not just how individuals code.

  • Git and GitHub workflows, branching, pull requests
  • Code review practice and etiquette
  • Agile/Scrum basics: standups, sprints, boards
  • Working in a squad on a shared codebase
MODULE 1.5 — MONTHS 5–6

Testing & Quality

Writing code that can be trusted — the habit that separates hobbyists from professionals.

  • Unit and integration testing
  • Debugging methodology
  • Linting, formatting, and code quality tooling
  • Intro to continuous integration (CI)
MONTHS 7–9

Applied AI & LLM foundations

Core fundamentals in hand, students layer on applied AI — starting with how to use large language models well, not just that they exist.

MODULE 2.1 — MONTH 7

LLM Literacy

What large language models actually are, what they're good at, and where they fail.

  • How LLMs are trained and how inference works, at a working level
  • Capabilities, limitations, and failure modes (hallucination, bias)
  • Responsible and ethical use of AI tools
MODULE 2.2 — MONTH 7

Prompting & Structured Outputs

Getting reliable, predictable results out of a fundamentally unpredictable system.

  • Prompt engineering techniques and few-shot examples
  • Structured output and JSON-mode generation
  • Iterating and testing prompts systematically
MODULE 2.3 — MONTH 8

Calling LLM APIs

Wiring real applications up to real model providers.

  • API integration and authentication
  • Streaming responses and handling latency
  • Token usage, cost tracking, and rate limits
MODULE 2.4 — MONTH 8

Retrieval-Augmented Generation

Grounding a model's answers in your own data instead of its training data.

  • Embeddings and vector similarity
  • Vector databases and indexing
  • Building a basic RAG pipeline end to end
MODULE 2.5 — MONTH 9

Classical ML Literacy

Enough traditional machine learning to know when it's the better tool than an LLM.

  • Supervised vs. unsupervised learning
  • Common algorithms: regression, classification, clustering
  • Choosing between rules, classical ML, and LLMs for a given problem
MONTHS 10–12

Agent building

From single model calls to systems that reason, act, and use tools on their own — under supervision.

MODULE 3.1 — MONTH 10

The Agent Loop

The core pattern behind every AI agent: reason, act, observe, repeat.

  • The reason-act-observe (ReAct) loop
  • Planning and task decomposition
  • When an agent should stop and ask for help
MODULE 3.2 — MONTH 10

Tool & Function Calling

Giving a model the ability to actually do things, not just talk about them.

  • Defining tool and function schemas
  • Executing and validating tool calls safely
  • Error handling when tools fail
MODULE 3.3 — MONTH 11

The Model Context Protocol (MCP)

The emerging standard for connecting agents to tools, data, and other systems.

  • MCP servers and clients
  • Connecting agents to external data sources and APIs
  • Building a simple MCP-compatible tool
MODULE 3.4 — MONTH 11

Memory & Orchestration

Making an agent remember context and coordinating multi-step workflows.

  • Short-term vs. long-term memory strategies
  • Orchestration frameworks and workflow design
  • State management across a multi-step task
MODULE 3.5 — MONTH 12

Multi-Agent Patterns

When one agent isn't enough — coordinating several toward a shared goal.

  • Supervisor and worker agent patterns
  • Agent-to-agent communication
  • Debugging multi-agent systems
MONTHS 13–16

Evaluation, production & Build Studio

Everything up to here was building the capability. This phase is about shipping it responsibly, then actually shipping it — to a real client.

MODULE 4.1 — MONTH 13

Agent Evaluation

Proving an agent actually works, before a client ever sees it.

  • Evaluation frameworks and test sets
  • Accuracy, consistency, and regression testing
  • Human review as part of the eval loop
MODULE 4.2 — MONTH 13

Guardrails & Safety

Keeping high-stakes actions safe, reversible, and accountable.

  • Human-in-the-loop approval for sensitive actions
  • Rate limits, permission boundaries, and kill switches
  • Logging and auditability
MODULE 4.3 — MONTH 14

Cost & Latency Management

Making a working system an affordable, responsive one.

  • Token and compute cost optimisation
  • Caching and response-time strategies
  • Model selection trade-offs (capability vs. cost)
MODULE 4.4 — MONTH 14

Deployment & Monitoring

Getting software out of a laptop and into the real world, reliably.

  • CI/CD pipelines
  • Production observability and alerting
  • Incident response basics
MODULE 4.5 — MONTHS 15–16

Build Studio Client Delivery

The culmination of the programme: a real client, a real deadline, a senior mentor's sign-off.

  • Scoping and managing a live client engagement
  • Working in a supervised delivery squad
  • Mentor review and sign-off before handover

Curious what the AI & Agents phases look like on their own? See the AI & Agents track on the homepage.