Free Course · 5 lessons · ~3.5 hours
Build a Market Research Bot
A hands-on introduction to using AI for serious market and company research. By the end you will have built a bot that researches a company end-to-end — and you will understand exactly why it must never make a trading decision for you.
Read this before you start
This course teaches you to build a bot that researches markets. It does not teach you to trade. The bot you build will never connect to a broker, never place an order, and never see your money.
Nothing in this course is financial advice, investment advice, or a recommendation to buy or sell any security. Any decisions you make about money are yours alone — and we strongly recommend you make them with a regulated financial adviser, not a chatbot. Markets carry real risk of loss.
Who this is for
You should take this course if you want to learn how AI helps with the research half of investing — summarising filings, parsing news, organising information across many sources — and if you want to understand the limits of using AI in a domain where being wrong costs real money.
The audience is broad. Finance professionals will recognise the workflow patterns. Software engineers will see how to architect a constrained, responsible AI assistant. Curious learners who have never read an annual report will get a working introduction to both AI and to how serious investors do research.
If you came hoping for "get rich with AI trading", this course will disappoint you, and we are glad. That is not what we teach.
What you will build
The scope and the rules
What this course is and is not. The legal context. The mandatory disclaimer. What a research bot can and cannot do for you.
Designing your bot's brief
The system prompt, the bot's "constitution", and the questions a good research bot should and should not answer.
Feeding it real material
Annual reports, earnings transcripts, news articles. How to give the bot information it does not already have, and how to make it cite sources.
Pattern recognition is not prediction
What AI is genuinely good at in research, and what it is not. Sentiment analysis, macro context, the temptation of false signals.
The risks, the regulations, the responsible use
Hallucination, false confidence, regulatory awareness, and how a serious investor actually uses tools like this. You will write a one-paragraph use policy for your bot.
What you will need
- About three to four hours. Each lesson is 25 to 50 minutes.
- A browser, and an account with either Claude.ai or ChatGPT. The free tier of either is enough.
- One or two real documents to research, in PDF or plain text. We will suggest specific public filings for each lesson. You can also use anything from your work.
How this fits with the full programme
This is the second of our free courses (after Build Your First AI Agent). It is standalone — you do not need the first course to take this one — but the two are complementary.
The deeper material is in our 180-ECTS Integrated AI Program. Tier 2 — C6 (Generative AI) goes into the techniques used here. Tier 3 Path B — B4 (Economics of AI) and B5 (AI Due Diligence, Investment and M&A) cover the responsible use of AI in financial decision-making. Tier 3 Path A — A2 (LLMs) and A4 (MLOps) cover how to deploy a research bot to production. The course wrap-up has the full list.