Skip to content
LeadAfrik
Beginner · Self-paced2026 Edition

AI for Economists & Analysts

What large language models, retrieval, and agents actually are — and what they mean for finance, economics, and analyst work. Twelve modules from foundations to deployment, taught the way a working analyst needs it: short on hype, long on what you can build with these tools tomorrow morning.

12

Modules

~10h 20m

Reading time

Beginner

Level

Self-paced

Format

§

Syllabus

  1. 01

    What AI is (and isn't) for an analyst

    Definitions, history, why now: scaling, data, compute. The supervised/unsupervised/RL split, and where LLMs sit.

    ~45 minModule 01
    →
  2. 02

    The machine-learning toolbox

    Regression, classification, clustering, trees, ensembles, neural networks — at a glance, with their best use cases.

    ~50 minModule 02
    →
  3. 03

    From OLS to neural networks

    The geometric and statistical lineage that connects a regression you understand to a neural network you don't.

    ~50 minModule 03
    →
  4. 04

    NLP foundations — tokens, embeddings, attention

    How language becomes numbers: tokenisation, word embeddings, the attention mechanism that powers every modern LLM.

    ~55 minModule 04
    →
  5. 05

    How large language models actually work

    GPT, Claude, Gemini — what the training run looks like, the pre-training / RLHF split, scaling laws, and why hallucinations happen.

    ~60 minModule 05
    →
  6. 06

    Prompt engineering for analysis

    The techniques that actually work: role, context, examples, step-by-step reasoning, and structured output.

    ~50 minModule 06
    →
  7. 07

    Retrieval-augmented generation (RAG)

    Build AI over your own documents — embeddings, vector databases, retrieval, and grounded answers that don't hallucinate.

    ~55 minModule 07
    →
  8. 08

    AI agents and tool use

    What it means for an LLM to call tools. Multi-step reasoning, planning loops, and the new agent frameworks.

    ~50 minModule 08
    →
  9. 09

    AI in finance and economics — real use cases

    Credit scoring, fraud, KYC, robo-advisors, trading, research automation. What works in production today, and what is still vaporware.

    ~55 minModule 09
    →
  10. 10

    AI for analyst productivity

    Cursor, Copilot, Claude, Gemini — the new analyst workflow. Prompts, code review, drafting, and the meta-skill of editing AI output.

    ~50 minModule 10
    →
  11. 11

    Evaluation, hallucination, and safety

    How to know if your AI is right. Eval sets, hallucination patterns, red-teaming, and the limits of guardrails.

    ~50 minModule 11
    →
  12. 12

    The economics, ethics, and geopolitics of AI

    Inference costs, jobs, regulation (EU AI Act, the US executive orders, Kenya's data-protection framework), and the chip-supply geopolitics.

    ~50 minModule 12
    →

How to use this course

Start with module 01 if the material is new; skip ahead if you have prior exposure. Each module is self-contained but the arc is sequential — the projects in the final module assume the toolkit from modules 1-11. Every module ends with key takeaways and a curated further-reading list with primary sources.