12. AI and Investing II: Using AI for Investment Signals

Course Positioning

A technical and research-oriented course on using AI to generate, test, and monitor investment signals responsibly.

Learning outcomes

Understand what an investment signal is, how it differs from a story, and why signals decay or fail.
Use AI to summarize filings, extract events, classify sentiment, detect themes, and monitor company-specific changes.
Build simple NLP, embedding, and time-series features from text, market, and fundamental data.
Design backtests that avoid look-ahead bias, survivorship bias, data snooping, overfitting, and unrealistic transaction assumptions.
Combine AI-generated signals with risk management, portfolio construction, and human judgment.
Create an ethical and compliant research workflow for personal investing or internal analysis.

Course Design Snapshot

Positioning: A technical and research-oriented course on using AI to generate, test, and monitor investment signals responsibly.
Audience: Quant-curious investors, analysts, data scientists, finance students, and professionals interested in systematic research.
Duration: 10 weeks, with optional coding track.
Prerequisites: Basic statistics, spreadsheets, and finance concepts. Python required for the technical track.
Format: Signal research labs, NLP on filings/news/transcripts, backtesting, risk controls, error analysis, and responsible-use guidance.

Expanded Topic-by-Topic Coverage

Module 1. Signal basics

Module focus: Signal basics: alpha, beta, factor exposure, noise, horizon, decay, correlation, and why prediction is hard. Primary live activity or lab: Turn five investing hypotheses into measurable signals.

Topics and coverage

alpha

What it means: define alpha clearly and connect it to the module focus: Signal basics: alpha, beta, factor exposure, noise, horizon, decay, correlation, and why prediction is hard.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

beta

What it means: define beta clearly and connect it to the module focus: Signal basics: alpha, beta, factor exposure, noise, horizon, decay, correlation, and why prediction is hard.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

factor exposure

What it means: define factor exposure clearly and connect it to the module focus: Signal basics: alpha, beta, factor exposure, noise, horizon, decay, correlation, and why prediction is hard.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

noise

What it means: define noise clearly and connect it to the module focus: Signal basics: alpha, beta, factor exposure, noise, horizon, decay, correlation, and why prediction is hard.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

horizon

What it means: define horizon clearly and connect it to the module focus: Signal basics: alpha, beta, factor exposure, noise, horizon, decay, correlation, and why prediction is hard.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

decay

What it means: define decay clearly and connect it to the module focus: Signal basics: alpha, beta, factor exposure, noise, horizon, decay, correlation, and why prediction is hard.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

correlation

What it means: define correlation clearly and connect it to the module focus: Signal basics: alpha, beta, factor exposure, noise, horizon, decay, correlation, and why prediction is hard.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

why prediction is hard

What it means: define why prediction is hard clearly and connect it to the module focus: Signal basics: alpha, beta, factor exposure, noise, horizon, decay, correlation, and why prediction is hard.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

Practice and evidence of learning

Learners complete or discuss: Turn five investing hypotheses into measurable signals.
Learners produce: Turn five investing hypotheses into measurable signals.
Instructor checks for accuracy, practical usefulness, clear assumptions, appropriate human review, and fit with the course audience.
Learners revise once after feedback so the module contributes to the final project, portfolio, or playbook.

Minimum coverage before moving on

Learners can explain the module vocabulary without relying on tool-generated text.
Learners have seen one worked example, one hands-on application, and one limitation or failure case.
Learners know what must be verified, what data must be protected, and who remains accountable for the output.

Module 2. Data sources and pitfalls

Module focus: Data sources and pitfalls: prices, fundamentals, filings, transcripts, news, social data, alternative data, and licensing. Primary live activity or lab: Create a data inventory and bias checklist.

Topics and coverage

prices

What it means: define prices clearly and connect it to the module focus: Data sources and pitfalls: prices, fundamentals, filings, transcripts, news, social data, alternative data, and licensing.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

fundamentals

What it means: define fundamentals clearly and connect it to the module focus: Data sources and pitfalls: prices, fundamentals, filings, transcripts, news, social data, alternative data, and licensing.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

filings

What it means: define filings clearly and connect it to the module focus: Data sources and pitfalls: prices, fundamentals, filings, transcripts, news, social data, alternative data, and licensing.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

transcripts

What it means: define transcripts clearly and connect it to the module focus: Data sources and pitfalls: prices, fundamentals, filings, transcripts, news, social data, alternative data, and licensing.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

news

What it means: define news clearly and connect it to the module focus: Data sources and pitfalls: prices, fundamentals, filings, transcripts, news, social data, alternative data, and licensing.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

What it means: connect social data to the data lifecycle from source and structure through analysis, interpretation, and decision-making.
What to cover: source reliability, missing or biased data, leakage, assumptions, calculations, and the difference between correlation and decision-ready evidence.
Demonstration: walk through a small dataset or example table and mark the checks required before trusting the result.
Evidence of learning: learners produce a short analysis note that includes assumptions, limitations, and verification steps.

alternative data

What it means: connect alternative data to the data lifecycle from source and structure through analysis, interpretation, and decision-making.
What to cover: source reliability, missing or biased data, leakage, assumptions, calculations, and the difference between correlation and decision-ready evidence.
Demonstration: walk through a small dataset or example table and mark the checks required before trusting the result.
Evidence of learning: learners produce a short analysis note that includes assumptions, limitations, and verification steps.

licensing

What it means: define licensing clearly and connect it to the module focus: Data sources and pitfalls: prices, fundamentals, filings, transcripts, news, social data, alternative data, and licensing.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

Practice and evidence of learning

Learners complete or discuss: Create a data inventory and bias checklist.
Learners produce: Create a data inventory and bias checklist.
Instructor checks for accuracy, practical usefulness, clear assumptions, appropriate human review, and fit with the course audience.
Learners revise once after feedback so the module contributes to the final project, portfolio, or playbook.

Minimum coverage before moving on

Learners can explain the module vocabulary without relying on tool-generated text.
Learners have seen one worked example, one hands-on application, and one limitation or failure case.
Learners know what must be verified, what data must be protected, and who remains accountable for the output.

Module 3. NLP for finance

Module focus: NLP for finance: tokenization, sentiment, topic modeling, embeddings, entity extraction, and document comparison. Primary live activity or lab: Analyze annual reports or earnings calls for risk themes.

Topics and coverage

tokenization

What it means: define tokenization clearly and connect it to the module focus: NLP for finance: tokenization, sentiment, topic modeling, embeddings, entity extraction, and document comparison.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

sentiment

What it means: define sentiment clearly and connect it to the module focus: NLP for finance: tokenization, sentiment, topic modeling, embeddings, entity extraction, and document comparison.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

topic modeling

What it means: place topic modeling inside the AI system stack so learners know what problem it solves and what tradeoffs it introduces.
What to cover: inputs, outputs, system boundaries, evaluation criteria, cost or latency implications, and common failure cases.
Demonstration: use a diagram, small code sample, worksheet, or tool trace to make the mechanism visible.
Evidence of learning: learners compare two approaches and explain which one they would choose for a realistic constraint.

embeddings

What it means: explain how embeddings changes the interaction between human intent, model behavior, external information, and final output.
What to cover: inputs, constraints, examples, output format, grounding, iteration, failure modes, and when a human must intervene.
Demonstration: show a weak attempt, a stronger structured attempt, and a reviewed final version with explicit checks.
Evidence of learning: learners create a reusable prompt, schema, retrieval note, or workflow pattern and test it on at least two examples.

entity extraction

What it means: define entity extraction clearly and connect it to the module focus: NLP for finance: tokenization, sentiment, topic modeling, embeddings, entity extraction, and document comparison.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

document comparison

What it means: define document comparison clearly and connect it to the module focus: NLP for finance: tokenization, sentiment, topic modeling, embeddings, entity extraction, and document comparison.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

Practice and evidence of learning

Learners complete or discuss: Analyze annual reports or earnings calls for risk themes.
Learners produce: Analyze annual reports or earnings calls for risk themes.
Instructor checks for accuracy, practical usefulness, clear assumptions, appropriate human review, and fit with the course audience.
Learners revise once after feedback so the module contributes to the final project, portfolio, or playbook.

Minimum coverage before moving on

Learners can explain the module vocabulary without relying on tool-generated text.
Learners have seen one worked example, one hands-on application, and one limitation or failure case.
Learners know what must be verified, what data must be protected, and who remains accountable for the output.

Module 4. LLMs for research workflows

Module focus: LLMs for research workflows: summarization, question answering, extraction, thesis tracking, and hallucination controls. Primary live activity or lab: Build a filing summarizer with citation and verification requirements.

Topics and coverage

summarization

What it means: define summarization clearly and connect it to the module focus: LLMs for research workflows: summarization, question answering, extraction, thesis tracking, and hallucination controls.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

question answering

What it means: define question answering clearly and connect it to the module focus: LLMs for research workflows: summarization, question answering, extraction, thesis tracking, and hallucination controls.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

extraction

What it means: define extraction clearly and connect it to the module focus: LLMs for research workflows: summarization, question answering, extraction, thesis tracking, and hallucination controls.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

thesis tracking

What it means: define thesis tracking clearly and connect it to the module focus: LLMs for research workflows: summarization, question answering, extraction, thesis tracking, and hallucination controls.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

hallucination controls

What it means: define hallucination controls clearly and connect it to the module focus: LLMs for research workflows: summarization, question answering, extraction, thesis tracking, and hallucination controls.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

Practice and evidence of learning

Learners complete or discuss: Build a filing summarizer with citation and verification requirements.
Learners produce: Build a filing summarizer with citation and verification requirements.
Instructor checks for accuracy, practical usefulness, clear assumptions, appropriate human review, and fit with the course audience.
Learners revise once after feedback so the module contributes to the final project, portfolio, or playbook.

Minimum coverage before moving on

Learners can explain the module vocabulary without relying on tool-generated text.
Learners have seen one worked example, one hands-on application, and one limitation or failure case.
Learners know what must be verified, what data must be protected, and who remains accountable for the output.

Module 5. Event detection

Module focus: Event detection: product launches, guidance changes, executive changes, legal risk, supply-chain changes, and capex plans. Primary live activity or lab: Create an event extraction schema and run it on sample text.

Topics and coverage

product launches

What it means: define product launches clearly and connect it to the module focus: Event detection: product launches, guidance changes, executive changes, legal risk, supply-chain changes, and capex plans.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

guidance changes

What it means: define guidance changes clearly and connect it to the module focus: Event detection: product launches, guidance changes, executive changes, legal risk, supply-chain changes, and capex plans.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

executive changes

What it means: define executive changes clearly and connect it to the module focus: Event detection: product launches, guidance changes, executive changes, legal risk, supply-chain changes, and capex plans.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

legal risk

What it means in this course: define legal risk in operational terms, not as an abstract principle.
What to cover: sensitive data boundaries, affected stakeholders, approval paths, documentation, and what Quant-curious investors, analysts, data scientists, finance students, and professionals interested in systematic research must never delegate blindly to AI.
Use case: present one acceptable use, one borderline use, and one prohibited use, then ask learners to justify the classification.
Evidence of learning: learners add a risk control, review step, or escalation rule to their course project.

supply-chain changes

What it means: define supply-chain changes clearly and connect it to the module focus: Event detection: product launches, guidance changes, executive changes, legal risk, supply-chain changes, and capex plans.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

capex plans

What it means: define capex plans clearly and connect it to the module focus: Event detection: product launches, guidance changes, executive changes, legal risk, supply-chain changes, and capex plans.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

Practice and evidence of learning

Learners complete or discuss: Create an event extraction schema and run it on sample text.
Learners produce: Create an event extraction schema and run it on sample text.
Instructor checks for accuracy, practical usefulness, clear assumptions, appropriate human review, and fit with the course audience.
Learners revise once after feedback so the module contributes to the final project, portfolio, or playbook.

Minimum coverage before moving on

Learners can explain the module vocabulary without relying on tool-generated text.
Learners have seen one worked example, one hands-on application, and one limitation or failure case.
Learners know what must be verified, what data must be protected, and who remains accountable for the output.

Module 6. Time-series and features

Module focus: Time-series and features: returns, volatility, momentum, mean reversion, revisions, seasonality, and feature leakage. Primary live activity or lab: Build simple price and fundamental features in Python or spreadsheet.

Topics and coverage

returns

What it means: define returns clearly and connect it to the module focus: Time-series and features: returns, volatility, momentum, mean reversion, revisions, seasonality, and feature leakage.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

volatility

What it means: define volatility clearly and connect it to the module focus: Time-series and features: returns, volatility, momentum, mean reversion, revisions, seasonality, and feature leakage.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

momentum

What it means: define momentum clearly and connect it to the module focus: Time-series and features: returns, volatility, momentum, mean reversion, revisions, seasonality, and feature leakage.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

mean reversion

What it means: define mean reversion clearly and connect it to the module focus: Time-series and features: returns, volatility, momentum, mean reversion, revisions, seasonality, and feature leakage.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

revisions

What it means: define revisions clearly and connect it to the module focus: Time-series and features: returns, volatility, momentum, mean reversion, revisions, seasonality, and feature leakage.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

seasonality

What it means: define seasonality clearly and connect it to the module focus: Time-series and features: returns, volatility, momentum, mean reversion, revisions, seasonality, and feature leakage.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

feature leakage

What it means: connect feature leakage to the data lifecycle from source and structure through analysis, interpretation, and decision-making.
What to cover: source reliability, missing or biased data, leakage, assumptions, calculations, and the difference between correlation and decision-ready evidence.
Demonstration: walk through a small dataset or example table and mark the checks required before trusting the result.
Evidence of learning: learners produce a short analysis note that includes assumptions, limitations, and verification steps.

Practice and evidence of learning

Learners complete or discuss: Build simple price and fundamental features in Python or spreadsheet.
Learners produce: Build simple price and fundamental features in Python or spreadsheet.
Instructor checks for accuracy, practical usefulness, clear assumptions, appropriate human review, and fit with the course audience.
Learners revise once after feedback so the module contributes to the final project, portfolio, or playbook.

Minimum coverage before moving on

Learners can explain the module vocabulary without relying on tool-generated text.
Learners have seen one worked example, one hands-on application, and one limitation or failure case.
Learners know what must be verified, what data must be protected, and who remains accountable for the output.

Module 7. Backtesting discipline

Module focus: Backtesting discipline: train/test periods, walk-forward validation, transaction costs, turnover, slippage, and benchmark choice. Primary live activity or lab: Run a toy backtest and then identify every unrealistic assumption.

Topics and coverage

train/test periods

What it means: define train/test periods clearly and connect it to the module focus: Backtesting discipline: train/test periods, walk-forward validation, transaction costs, turnover, slippage, and benchmark choice.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

walk-forward validation

What it means: define walk-forward validation clearly and connect it to the module focus: Backtesting discipline: train/test periods, walk-forward validation, transaction costs, turnover, slippage, and benchmark choice.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

transaction costs

What it means: place transaction costs inside the AI system stack so learners know what problem it solves and what tradeoffs it introduces.
What to cover: inputs, outputs, system boundaries, evaluation criteria, cost or latency implications, and common failure cases.
Demonstration: use a diagram, small code sample, worksheet, or tool trace to make the mechanism visible.
Evidence of learning: learners compare two approaches and explain which one they would choose for a realistic constraint.

turnover

What it means: define turnover clearly and connect it to the module focus: Backtesting discipline: train/test periods, walk-forward validation, transaction costs, turnover, slippage, and benchmark choice.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

slippage

What it means: define slippage clearly and connect it to the module focus: Backtesting discipline: train/test periods, walk-forward validation, transaction costs, turnover, slippage, and benchmark choice.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

benchmark choice

What it means: define benchmark choice clearly and connect it to the module focus: Backtesting discipline: train/test periods, walk-forward validation, transaction costs, turnover, slippage, and benchmark choice.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

Practice and evidence of learning

Learners complete or discuss: Run a toy backtest and then identify every unrealistic assumption.
Learners produce: Run a toy backtest and then identify every unrealistic assumption.
Instructor checks for accuracy, practical usefulness, clear assumptions, appropriate human review, and fit with the course audience.
Learners revise once after feedback so the module contributes to the final project, portfolio, or playbook.

Minimum coverage before moving on

Learners can explain the module vocabulary without relying on tool-generated text.
Learners have seen one worked example, one hands-on application, and one limitation or failure case.
Learners know what must be verified, what data must be protected, and who remains accountable for the output.

Module 8. Portfolio construction

Module focus: Portfolio construction: ranking, weighting, risk parity intuition, sector exposure, drawdowns, and stress testing. Primary live activity or lab: Convert signals into a paper portfolio and analyze risk exposure.

Topics and coverage

ranking

What it means: define ranking clearly and connect it to the module focus: Portfolio construction: ranking, weighting, risk parity intuition, sector exposure, drawdowns, and stress testing.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

weighting

What it means: define weighting clearly and connect it to the module focus: Portfolio construction: ranking, weighting, risk parity intuition, sector exposure, drawdowns, and stress testing.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

risk parity intuition

What it means in this course: define risk parity intuition in operational terms, not as an abstract principle.
What to cover: sensitive data boundaries, affected stakeholders, approval paths, documentation, and what Quant-curious investors, analysts, data scientists, finance students, and professionals interested in systematic research must never delegate blindly to AI.
Use case: present one acceptable use, one borderline use, and one prohibited use, then ask learners to justify the classification.
Evidence of learning: learners add a risk control, review step, or escalation rule to their course project.

sector exposure

What it means: define sector exposure clearly and connect it to the module focus: Portfolio construction: ranking, weighting, risk parity intuition, sector exposure, drawdowns, and stress testing.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

drawdowns

What it means: define drawdowns clearly and connect it to the module focus: Portfolio construction: ranking, weighting, risk parity intuition, sector exposure, drawdowns, and stress testing.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

stress testing

What it means: define stress testing clearly and connect it to the module focus: Portfolio construction: ranking, weighting, risk parity intuition, sector exposure, drawdowns, and stress testing.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

Practice and evidence of learning

Learners complete or discuss: Convert signals into a paper portfolio and analyze risk exposure.
Learners produce: Convert signals into a paper portfolio and analyze risk exposure.
Instructor checks for accuracy, practical usefulness, clear assumptions, appropriate human review, and fit with the course audience.
Learners revise once after feedback so the module contributes to the final project, portfolio, or playbook.

Minimum coverage before moving on

Learners can explain the module vocabulary without relying on tool-generated text.
Learners have seen one worked example, one hands-on application, and one limitation or failure case.
Learners know what must be verified, what data must be protected, and who remains accountable for the output.

Module 9. Monitoring and signal governance

Module focus: Monitoring and signal governance: drift, decay, model updates, audit logs, compliance, and human review. Primary live activity or lab: Design a signal dashboard and review cadence.

Topics and coverage

drift

What it means: define drift clearly and connect it to the module focus: Monitoring and signal governance: drift, decay, model updates, audit logs, compliance, and human review.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

decay

What it means: define decay clearly and connect it to the module focus: Monitoring and signal governance: drift, decay, model updates, audit logs, compliance, and human review.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

model updates

What it means: place model updates inside the AI system stack so learners know what problem it solves and what tradeoffs it introduces.
What to cover: inputs, outputs, system boundaries, evaluation criteria, cost or latency implications, and common failure cases.
Demonstration: use a diagram, small code sample, worksheet, or tool trace to make the mechanism visible.
Evidence of learning: learners compare two approaches and explain which one they would choose for a realistic constraint.

audit logs

What it means: define audit logs clearly and connect it to the module focus: Monitoring and signal governance: drift, decay, model updates, audit logs, compliance, and human review.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

compliance

What it means in this course: define compliance in operational terms, not as an abstract principle.
What to cover: sensitive data boundaries, affected stakeholders, approval paths, documentation, and what Quant-curious investors, analysts, data scientists, finance students, and professionals interested in systematic research must never delegate blindly to AI.
Use case: present one acceptable use, one borderline use, and one prohibited use, then ask learners to justify the classification.
Evidence of learning: learners add a risk control, review step, or escalation rule to their course project.

human review

What it means: define human review clearly and connect it to the module focus: Monitoring and signal governance: drift, decay, model updates, audit logs, compliance, and human review.
What to cover: the core concept, why it matters, what good usage looks like, and where learners are likely to misunderstand it.
Demonstration: give one simple example, one realistic example, and one failure or limitation example.
Evidence of learning: learners explain the topic in their own words and apply it to a small artifact or decision.

Practice and evidence of learning

Learners complete or discuss: Design a signal dashboard and review cadence.
Learners produce: Design a signal dashboard and review cadence.
Instructor checks for accuracy, practical usefulness, clear assumptions, appropriate human review, and fit with the course audience.
Learners revise once after feedback so the module contributes to the final project, portfolio, or playbook.

Minimum coverage before moving on

Learners can explain the module vocabulary without relying on tool-generated text.
Learners have seen one worked example, one hands-on application, and one limitation or failure case.
Learners know what must be verified, what data must be protected, and who remains accountable for the output.

Module 10. Capstone research sprint

Module focus: Capstone research sprint: signal thesis, dataset, feature, backtest, risk analysis, and research memo. Primary live activity or lab: Present a signal research memo with caveats and next tests.

Topics and coverage

signal thesis

What it means: connect signal thesis to the data lifecycle from source and structure through analysis, interpretation, and decision-making.
What to cover: source reliability, missing or biased data, leakage, assumptions, calculations, and the difference between correlation and decision-ready evidence.
Demonstration: walk through a small dataset or example table and mark the checks required before trusting the result.
Evidence of learning: learners produce a short analysis note that includes assumptions, limitations, and verification steps.

dataset

What it means: connect dataset to the data lifecycle from source and structure through analysis, interpretation, and decision-making.
What to cover: source reliability, missing or biased data, leakage, assumptions, calculations, and the difference between correlation and decision-ready evidence.
Demonstration: walk through a small dataset or example table and mark the checks required before trusting the result.
Evidence of learning: learners produce a short analysis note that includes assumptions, limitations, and verification steps.

feature

What it means: connect feature to the data lifecycle from source and structure through analysis, interpretation, and decision-making.
What to cover: source reliability, missing or biased data, leakage, assumptions, calculations, and the difference between correlation and decision-ready evidence.
Demonstration: walk through a small dataset or example table and mark the checks required before trusting the result.
Evidence of learning: learners produce a short analysis note that includes assumptions, limitations, and verification steps.

backtest

What it means: connect backtest to the data lifecycle from source and structure through analysis, interpretation, and decision-making.
What to cover: source reliability, missing or biased data, leakage, assumptions, calculations, and the difference between correlation and decision-ready evidence.
Demonstration: walk through a small dataset or example table and mark the checks required before trusting the result.
Evidence of learning: learners produce a short analysis note that includes assumptions, limitations, and verification steps.

risk analysis

What it means in this course: define risk analysis in operational terms, not as an abstract principle.
What to cover: sensitive data boundaries, affected stakeholders, approval paths, documentation, and what Quant-curious investors, analysts, data scientists, finance students, and professionals interested in systematic research must never delegate blindly to AI.
Use case: present one acceptable use, one borderline use, and one prohibited use, then ask learners to justify the classification.
Evidence of learning: learners add a risk control, review step, or escalation rule to their course project.

research memo

What it means: show where research memo appears in the learner's real workflow and which parts are judgment-heavy versus draftable.
What to cover: current workflow, pain points, AI-assisted steps, human review checkpoints, quality standard, and ownership of the final decision.
Demonstration: convert one messy real-world input into a structured brief, draft, analysis, checklist, or next action.
Evidence of learning: learners produce a reusable template or playbook entry that can be used after the course.

Practice and evidence of learning

Learners complete or discuss: Present a signal research memo with caveats and next tests.
Learners produce: Present a signal research memo with caveats and next tests.
Instructor checks for accuracy, practical usefulness, clear assumptions, appropriate human review, and fit with the course audience.
Learners revise once after feedback so the module contributes to the final project, portfolio, or playbook.

Minimum coverage before moving on

Learners can explain the module vocabulary without relying on tool-generated text.
Learners have seen one worked example, one hands-on application, and one limitation or failure case.
Learners know what must be verified, what data must be protected, and who remains accountable for the output.

Core labs and builds

Filings comparison lab: detect changes in risk factors or AI strategy language over time.
Earnings transcript lab: extract sentiment, themes, capex discussion, and uncertainty language.
Backtest pitfalls lab: intentionally overfit a strategy, then repair the research design.
Signal dashboard lab: combine text signal, price signal, risk metrics, and human review notes.

Capstone

Build and document one AI-assisted investment signal, such as AI-capex intensity, earnings-call uncertainty, patent/topic momentum, customer concentration language, hiring signal, or news-event sentiment. The final memo must include hypothesis, data, feature construction, backtest, limitations, risk controls, and why it may fail live.

Assessment design

Signal hypothesis quality and measurability.
Data hygiene and bias controls.
Backtest realism and error analysis.
Final research memo with clear limitations and no exaggerated claims.

Recommended tools and datasets

Python, pandas, yfinance or licensed data source, SEC/annual reports, earnings-call transcripts, NLP libraries, embeddings, vector search, spreadsheets, backtesting templates, dashboard tool.

Instructor notes

The most important lesson is skepticism. AI can accelerate research, but it can also accelerate overfitting, confirmation bias, and false confidence.
Selected reference signals for curriculum design.
These sources were used as background signals while shaping course emphasis around AI competency, risk management, workforce change, enterprise adoption, hardware trends, and investing safeguards.
UNESCO AI Competency Framework for Students and Teachers: AI literacy, responsible use, education competencies, and curricular framing.
NIST AI Risk Management Framework 1.0: Risk management, trustworthy AI, governance, impact analysis, and lifecycle thinking.
OECD AI Principles, updated 2024: Human-centered, trustworthy, rights-respecting AI governance principles.
Stanford HAI AI Index 2025: Trends in AI capability, hardware efficiency, inference costs, policy, research, and adoption.
World Economic Forum Future of Jobs Report 2025: Workforce transformation, skill disruption, and employer upskilling priorities.
ILO Generative AI and Jobs 2025 update: Task-level exposure analysis and labor market framing.
McKinsey State of AI Global Survey 2025: Enterprise AI scaling, value capture, operating practices, and agentic AI adoption challenges.
Google Cloud TPU documentation: Specialized AI accelerators and cloud AI hardware framing.
SEC, NASAA, and FINRA investor alerts on AI and investment fraud: Risk warnings for AI-related investing education and AI-washing awareness.
Suggested packaging into products.
School technical ladder: Course 1 -> Course 2 -> holiday bootcamp -> science fair/AI project showcase.
College technical certificate: Course 3 plus optional add-ons from Courses 7 and 9.
Professional AI builder certificate: Course 4 followed by an applied project clinic.
AI leadership program: Course 5 plus selected modules from Courses 8, 9, and 10.
Public AI literacy masterclass: Course 6 as a broad entry course for educated general audiences.
Business owner program: Course 10 plus templates, SOP packs, and monthly implementation office hours.
AI investing education series: Course 11 for fundamentals and Course 12 only for learners who understand risk and data limitations.

Instructor Build Checklist

Prepare one short demo for each module and one learner activity that creates a saved artifact.
Prepare examples that match the audience, local context, and likely tools learners can access.
Add a verification step to every AI-generated output: factual check, source check, data sensitivity check, and quality review.
Keep a running portfolio folder so each module contributes to the final project or learner playbook.
Reserve time for reflection on what the learner did, what AI did, what was checked, and what remains uncertain.