3–5
Documented projects
Each with a short write-up: the problem, your approach, the trade-offs, the result, and how it was evaluated.
A portfolio that translates into a job
You don't build throwaway practice projects. You contribute to real, in-progress work — so what you graduate with is evidence you can do the job on day one.
What “real” means
Real project work, reviewed like production
Every build in a CapabilityNext programme is a genuine project: real requirements, real data, real constraints, and a real review. It's assessed the way a senior engineer reviews a pull request, not marked like a worksheet.
- Briefs come from actual work: internal tools, data problems, model deployments.
- You handle the messy parts: missing data, ambiguous specs, latency and budget limits.
- Every submission gets written mentor feedback plus a live review.
- “It runs” isn't the bar. “You can explain and defend it” is.
What a review looks at
CorrectnessDoes it work on data it has not seen?
ClarityCan another engineer read, run and extend it?
Trade-offsWhy this approach and not the alternatives?
CostWhat does it cost to run, and where would it break?
What you leave with
The portfolio, unpacked
1
Presented capstone
A complete project on a real brief, demoed at a showcase. The piece you walk an interviewer through.
1
Mentor reference
A practitioner who reviewed your work and can speak to how you think, not just that you attended.
It's structured to map onto what technical interviews actually ask about: architecture decisions, evaluation, cost, and what you'd change.
How it flows
From brief to interview
- 1Start
A real brief
Requirements and data that look like the job.
- 2Build
Ship the project
Handle the messy parts, keep a write-up as you go.
- 3Review
Mentor feedback
Written notes plus a live review of your choices.
- 4Show
Present the capstone
Demo it at a showcase and defend the trade-offs.
From portfolio to payroll
Each programme is built around a live role
Before a programme runs, we work backwards from a real hiring need: the responsibilities, the tools, and what those interviews screen for. The curriculum, projects and capstone are shaped to that profile, so on completion you're ready for a specific kind of role and can start applying immediately. See the salary and skills map for each role →
Interview preparation
Tuned to AI roles: technical rounds, take-homes, and how to present your portfolio.
CV and portfolio review
A mentor goes through both with you before you apply.
Introductions
Where we have them, we make them.
An honest caveat: we don't guarantee any individual offer. That still comes down to your interviews.
Student work
A sample of what graduates ship
The kind of work that comes out of a cohort. These are illustrative examples of the standard we aim for, not named student projects.
GenAI & LLM
Policy assistant with source citations
RAG assistant over a long handbook, scored against a fixed test set of questions and held to a fixed cost per query.
Data & MLOps
Churn model, deployed and monitored
End-to-end pipeline: feature store, containerised serving, drift alerts and an automated retraining trigger.
Computer Vision
Document layout extractor
Detection + OCR pipeline turning scanned invoices into structured JSON, evaluated on a held-out set.
Feedback you can act on
What written mentor feedback looks like
An example of the format every submission gets. Student stories will appear here once graduates have agreed to share them.
Example
Capstone review: retrieval assistantMentor notes
Strong
Clear separation of ingestion and query code, and a test set you built before tuning anything.
Improve
Chunk size was chosen once and never compared. Try two alternatives and report retrieval hit-rate for each.
Next
Add a “can't answer from these documents” path and show how it behaves on out-of-scope questions.
From portfolio to payroll
See where a course path leads
Tick the courses you have taken to see how ready you would be for an AI/ML engineer role, and what it typically pays.
More roles on the career & salary map.
Your portfolio
Build the portfolio that gets you hired
Tell us the role you're aiming for and we'll map it to a track.
Talk to an advisor