Skip to main content

Capability

Hands-on AI Project Programmes

Programmes in which teams build a working AI project on a real problem, following the CRISP-DM method from business understanding to evaluation and presentation. Participants leave with a result they built themselves and a method they can repeat.

People learn AI fastest by building with it. We structure that building so it produces something useful and a method that lasts.

01Architecture

Reference architecture

A typical arrangement of the components. Every implementation is adapted to your systems, data-residency and security requirements.

Business question

  • Problem framing
  • Success criteria
Approved data and tools

Lab environment

  • NotebooksJupyter · Google Colab
  • Sample datasetspandas
  • Models and librariesscikit-learn · Hugging Face
CRISP-DM build cycles

Team project

  • Working prototypeStreamlit
  • Evaluation against criteria
Presentation

Review and handover

  • Written feedback
  • Code and notebooks handed overGitHub
Components shown by layer, top to bottom, with what flows between them. Technologies are examples, chosen per project.

02Components

What this capability covers

  1. 01

    Problem framing

    Teams start from a business question and agree what a useful result would look like.

  2. 02

    Prepared lab environment

    Notebooks, sample datasets and tools set up in advance, so session time goes to building.

  3. 03

    Mentored build cycles

    Short build and review rounds with feedback on data preparation, modelling and evaluation.

  4. 04

    Presentation and review

    Each team presents its result and its limits, and receives written feedback.

03Implementation

How we implement it

Each programme follows CRISP-DM: business understanding, data understanding, data preparation, modelling, evaluation and presentation. We prepare the lab environment and datasets before the first session and agree with you which data may be used, so teams work on realistic material without exposing sensitive records.

  • Small teams combine domain knowledge with technical skills.
  • Mentors review progress at every phase, not only at the end.
  • Projects, notebooks and feedback are handed over to participants and, where agreed, to the sponsoring organisation.

04Deliverables

What you receive

  • Prepared lab environment and datasets

  • Project brief per team

  • Working prototypes with their evaluation

  • Feedback and programme report

Technology

Technologies we work with

  • Python
  • Jupyter
  • Google Colab
  • pandas
  • scikit-learn
  • Hugging Face
  • Streamlit
  • GitHub

Chosen per project and aligned with your existing standards. Listed as technologies we use, not as partnerships or endorsements.

05Use cases

Typical applications

Patterns this capability is designed for, not a list of delivered projects.

  1. 01

    Open cohort programmes

    Professionals from different organisations each build a project on a problem from their own field.

  2. 02

    In-house project sprints

    Mixed business and technical teams build a first prototype on an approved internal use case.

  3. 03

    Early-career programmes

    Structured project work that gives new talent practical AI experience.

06Evidence

Related work

Projects and programmes in which this capability played a part, described with the role NLAI actually had.

  • Training programme

    AI Project Building Course — Cohort 9

    Cohort 9 of the AI Project Building Course was a three-day, in-person programme that NLAI delivered in a training partnership with EMC (Educational Mastar Central).

    Read more

07Related

More in Training & Capability Building

  • Role-based AI Curriculum Design

    Learning paths for executives, managers, specialists and technical teams, built from a skills analysis and your own processes rather than generic examples.

Next step

Scope Hands-on AI Project Programmes for your organisation

Share your process, data and constraints. We will propose an approach, the evidence to collect first and a realistic plan.