Quantitative analysts, financial developers, researchers, and technically minded investment professionals.
PROGRAM 02 / FROM FINANCIAL DATA TO DECISION SUPPORT
Financial AI Engineering
Build research and analytical workflows around financial data. Connect documents, quantitative methods, and agents with explicit controls.
- Financial data
- Research agents
- Risk & backtesting
- Document intelligence
Python fundamentals and basic statistics. Familiarity with financial statements or market data is useful.
A research workbench combining financial data, document evidence, risk analysis, and a human-reviewed research agent.
THE CURRICULUM
From foundations
to a working project.
Open a module to see its agenda and practical outcome. The learning path builds across three stages.
STAGE 01
Data & research foundations
01Financial data contracts
- Model instruments, currencies, timestamps, and observations.
- Validate provider responses and missing values.
- Keep data provenance and publication time visible.
02Financial documents as evidence
- Extract text and tables from reports and presentations.
- Preserve document dates, units, and source locations.
- Compare extracted figures with the original document.
03Research retrieval and grounded answers
- Design retrieval for filings, transcripts, and research notes.
- Combine metadata filters, keyword search, and embeddings.
- Distinguish source evidence from model interpretation.
04Feature design and statistical baselines
- Explore distributions, missing data, and unstable relationships.
- Create features without using future information.
- Compare simple baselines before adding model complexity.
STAGE 02
Quantitative analysis
05Risk and portfolio analysis
- Measure exposure, drawdown, concentration, and volatility.
- Explore scenario analysis and portfolio constraints.
- Explain model assumptions and estimation uncertainty.
06Time series and forecasting
- Use time-aware splits and rolling evaluation.
- Compare statistical and machine learning forecasts.
- Test sensitivity to regime changes and data revisions.
07Backtesting without shortcuts
- Model costs, execution delays, and available information.
- Separate parameter selection from final evaluation.
- Use walk-forward analysis and investigate fragile results.
08Typed financial research agents
- Connect analytical tools to a PydanticAI agent.
- Separate numerical calculations from generated narrative.
- Require review before external or financial actions.
STAGE 03
Controlled financial workflows
09Relationships and graph-assisted research
- Represent relationships between companies and documents.
- Combine relationship queries with document retrieval.
- Check where a graph adds useful evidence.
10Model adaptation and financial language
- Compare prompting, retrieval, and fine-tuning.
- Prepare a small, well-defined evaluation dataset.
- Assess local model deployment and operating costs.
11Integrations, monitoring, and controls
- Connect approved analytical tools through MCP.
- Trace sources, tool calls, errors, and model costs.
- Design permissions, review points, and recovery steps.
12Final project: financial research workbench
- Combine data validation, document evidence, and risk views.
- Demonstrate an agent-assisted research process.
- Present the evaluation, controls, and unresolved limitations.
FINANCIAL AI / WAITLIST
Your next step
starts here.
Join the waitlist for Financial AI Engineering. We will send the cohort dates, delivery schedule, and enrolment details when they are ready.
AED 6,999 full program fee
This waitlist is only for this program. You can join another program separately.
EXPLORE ANOTHER DIRECTION