APPLIED AI / DATA SCIENCE / CHANGE

AI & Data Science Portfolio

I find decisions that need better evidence, choose data that can be used responsibly, and turn analysis into a practical change people can own.

Explore the teaching models

THE THREAD THROUGH THE WORK

Discovery, evidence, action.

These projects grew from workplace questions: where procurement slows, how students experience digital services, what a feedback log can and cannot tell us, and whether software licences are being used well. Alongside the investigations, I built a more data-aware Business Analysis Capability: a structured service offer, project-specific delivery frameworks, an AI and data learning pathway, and a method for assessing opportunities before investing in a model. My MSc methods strengthened the analysis; internal consultancy and change leadership made the work usable.

Each case identifies its data basis and stage. Restricted real-data analyses, synthetic demonstrations and academic governance proposals have different kinds of evidence and different limits. The public pages contain no raw workplace records. The eight BA data assets behind the capability are described separately ↗

How I take a use case through delivery ↗

Follow a real question into the method.

Begin with the case studies below. To test the analytical ideas, visit the Smart Procurement Lab or Applied Analytics Lab. Both expose their fictional inputs, methods and limits alongside the outputs.

THE WORK, IN VIEW

What the diagrams and model outputs reveal.

Explore the BA operating model, persona playbook, controlled agent checks and actual plots from a synthetic ML notebook. Restricted workplace screens are represented by clearly labelled public reconstructions.

See the visual evidence

Synthetic purchase-order delay feature-importance chart

ACTUAL NOTEBOOK OUTPUT · FICTIONAL RECORDS

OPEN STUDY DESK

References for the work behind the work.

Explore the Python, data, visualisation and machine learning guides I used during my MSc. Each entry names who it helps, what to use it for, and where to open or save the publisher’s PDF.

Explore the free learning resources

MILESTONES / THE CAPABILITY IN PRACTICE

From a stronger foundation to a better decision.

These stages connect BA Capability leadership, MSc study and applied investigations. They describe distinct kinds of work rather than a single deployed platform.

01 / FOUNDATION

Make work legible

Led the BA Capability, a team of 10 analysts, and authored a catalogue of more than 100 activities mapped to eight delivery gates. Process, persona and data-requirements guidance made the service offer more explicit and reusable.

BA Capability & internal consultancy ↗ 02 / CAPABILITY

Designed a six-course, roughly 29-hour team pathway from data literacy to applied AI and Python, with seven group sessions proposed alongside self-study. My nine-module MSc is a separate personal qualification.

Module map ↗ 03 / REAL-DATA INSIGHT

Make evidence comparable

Harmonised real student-feedback exports across 2023/24–2025/26. Separately explored restricted feedback and software-usage sources, each for a defined decision and with its own limits.

Survey case ↗ 04 / SYNTHETIC TESTING

Test before relying

Used 5,000 curated synthetic purchase-order records to investigate four model questions. Related academic studies rehearsed staff and student experience methods without live prediction claims.

Capstone ↗ 05 / ASSURANCE

Put judgment beside the signal

Compared baselines, explained uncertainty and identified review owners. A promising prototype needs permission, real-world validation and adoption measures before operational use.

Delivery method ↗

TEAM & PROJECT MILESTONES

Building the conditions for useful analysis.

The artefacts below show how I translated individual learning into a capability standards and project methods. Their delivery stage is part of the story.

DELIVERED · CAPABILITY METHOD

A service offer analysts can explain

A structured activity catalogue, eight-gate alignment and process-modelling standard made work easier to scope, trace and review. A Jira board was prototyped as a further visibility step.

Explore the BA Capability ↗ AUTHORED · PROJECT FRAMEWORKS

Data questions at the right stage

I prepared tailored BA approaches for learner engagement, service management, assets and workplace operations: source, owner, quality, interface, reporting and adoption questions. The documents describe recommended work, not completed system outcomes.

See my approach ↗ TESTED · CONTROLLED POC

Agent feasibility with guardrails

An October 2025 guide reported 91% success across 11 controlled Copilot Studio checks. Synthetic examples and test identities supported the PoC; wider service readiness remained a separate decision.

Read the PoC ↗ DESIGNED · TEAM LEARNING

From individual study to shared fluency

A six-course pathway brought about 29 hours of material into a plan for seven three-hour team sessions plus self-study. The schedule and showcases supported adoption design; completion by every colleague is not claimed.

Explore the learning ↗

01 / APPLIED CASES

Workplace questions, tested methods

Open a case to see the question, dataset, approach, intended change and limits.

OPERATIONS · SYNTHETIC DEMONSTRATOR

Procurement intelligence

Real process discovery shaped four questions; curated synthetic records tested delay, approval, anomaly and carbon estimates in a notebook demonstrator.

Read the case ↗ STUDENT EXPERIENCE · RESTRICTED REAL DATA

Welcome Week insight

Three annual feedback cycles were cleaned, aligned and compared, with missing answers and proxy measures made visible.

Read the case ↗ PEOPLE ANALYTICS · RESTRICTED REAL DATA

BA feedback analytics

Text and thematic methods explored a positive-feedback log; selective recording makes individual performance inference unsafe.

Read the case ↗ TOOL GOVERNANCE · RESTRICTED REAL DATA

Software licence intelligence

Usage and licence fields were standardised to flag questions for adoption support and renewal review, with context before action.

Read the case ↗ PERSONAL FIELD NOTES · EXPLORATORY PYTHON

Chelsfield football lab

Mark’s club statistics prompted my 2025 Python notebook. The new browser-only player lab explores dated form, milestones, editable SWOTs and an honest walk-forward forecast check without copying his database.

Visit the football page ↗

02 / CONNECTED AREAS

Three places the capability travels.

01 / OPERATIONS

Process and procurement

Follow a problem from stakeholder mapping through a governed synthetic model demonstrator.

Explore operations ↗02 / PEOPLE & EDUCATION

Experience and early signals

Connect real service feedback with carefully bounded academic prediction exercises.

Explore the research ↗03 / CULTURE & LANGUAGE

Whose knowledge counts?

Yorùbá language governance and heritage ethics connect technical work to cultural stewardship.

Explore cultural AI ↗

03 / THE LEARNING ENGINE

From MSc learning to a team capability.

Watch my QMUL staff-apprenticeship interview: returning to mathematics, building confidence and bringing AI and data science learning into the team I led.

Nine connected modules, including the capstone, link technical methods to decisions, communication and responsible change. Explore selected studies and the evidence behind them.

Explore the module map

Browse 60 learning topics · How I would measure adoption and service value

Python code and notebooks for selected studies are privately stored and maintained by me. I’m happy to answer questions about the methods and their limits.

A CLOSER LOOK

Inspect the errors. Original capstone notebook output using synthetic records. False alarms and missed cases must be examined before any real-data trial. — Original capstone notebook output using synthetic records. False alarms and missed cases must be examined before any real-data trial.

What the model relied on. Original synthetic delay-model chart. Feature importance describes model behaviour; it does not establish causation. — Original synthetic delay-model chart. Feature importance describes model behaviour; it does not establish causation.

THE CONNECTIONMy broader approach ↗Tapestries of Yorùbá Civilization ↗Share a thought ↗

Try the Smart Procurement Lab.

Four explainable prototypes: PO delay · Approval bottlenecks · Anomaly review · Carbon impact. Change fictional inputs, compare scenarios and inspect measured synthetic-test results.

BUSINESS ANALYSIS & CONSULTANCY FOUNDATIONS

Earlier work. Enduring methods.

Explore non-AI delivery examples separately: requirements, service change, migration discovery and facilitation. These experiences underpin the AI and data portfolio.

Explore the foundational projects and two practical exercises · Six steps from process evidence to responsible change

APPLIED AI / REGIONAL FORECAST LAB

Explore a future. Inspect the model behind it.

Train a regional revenue model, inspect chronological backtests and adjust population, youth, GDP and resource scenarios.

Open the Regional Forecast Lab →

EXPLAINABLE AI & DATA / READ THE METHOD

Read the evidence, understand the method.

A guide to the charts and case evidence on this page. Exhibit captions remain the authority for data origin and the claims each visual supports.

Follow the reasoning.

SourceRead the exhibit’s evidence label
RepresentInspect units, categories and scales
CompareDistinguish observed results from demonstrations
ReviewCheck limitations before taking action

Explore the building blocks.

AI Capability4

AI/Data Concepts2

Data Science Pipeline1

Evaluation & XAI3

Python Package2

Responsible AI1

Visualisation1

Number of concepts in this guide, not performance or usage measures.

Explore the explanation matrix · 14 concepts

Concept categoryFind a term or example

14 concepts shown

On small screens, scroll the table sideways. “Learning context” explains a related idea; it does not mean that method runs on this page. Deep learning is the standard term for “deep ML”.

Technical term	Technical description / definition	In plain English	How this page uses it
Visual encodingVisualisation	Mapping values or categories to marks, position, length, colour and labels.Primary reference ↗	Make the comparison visible, with a scale and a clear label.	Published analysis Read axis labels, denominators and captions before comparing exhibits.
Synthetic dataAI/Data Concepts	Generated records used to represent scenarios rather than directly report observed individuals or events.Primary reference ↗	A practice dataset. Useful for demonstration, not proof of workplace impact.	Demonstration Exhibits labelled synthetic or reconstructed illustrate methods; they are not disclosed workplace records.
Descriptive analyticsAI/Data Concepts	Summaries of observed data using counts, rates, distributions or comparisons.Primary reference ↗	Tell us what the records show; this alone does not predict next week.	Published analysis Counts and trend summaries describe the stated records and period, not future performance.
Residuals and prediction errorEvaluation & XAI	A residual compares an observed value with its fitted or predicted value. Error metrics summarise those discrepancies.Primary reference ↗	How far did the estimate miss, and where does it miss most?	Learning context Prediction exhibits can compare observed and estimated values; a smooth chart alone is not validation.
Confusion matrix, precision and recallEvaluation & XAI	A confusion matrix counts predicted versus actual classes. Precision measures positive-result reliability; recall measures positive-case coverage.Primary reference ↗	Which alarms were right, which were false, and which cases were missed?	Learning context Where a confusion chart is shown, inspect false alarms and missed cases as well as correct predictions.
Data science pipelineData Science Pipeline	An ordered, reproducible sequence for acquiring, validating, preparing, analysing and evaluating data.Primary reference ↗	A chain from a question and source to an answer someone can check.	Published analysis Case studies connect the problem, evidence, preparation, method, evaluation and decision.
Explanation versus causal proofEvaluation & XAI	A model explanation describes selected behaviour or contributions; it does not establish causal effects or full reliability.Primary reference ↗	A reason for this output is useful, but it is not proof that the model is right.	Responsible use A persuasive visual does not prove causation or transferability.
Data minimisationResponsible AI	Limit collected, disclosed and retained information to what is needed for the stated purpose.Primary reference ↗	Use the smallest appropriate amount of information.	Responsible use Restricted source records stay private; approved aggregates and labelled reconstructions support public learning.
pandasPython Package	A Python library for labelled tabular data manipulation and analysis.Primary reference ↗	A programmable workbench for cleaning and comparing tables.	Learning context A common Python option for tabular preparation; its inclusion is not a claim that this browser runs pandas.
MatplotlibPython Package	A Python plotting library for static, animated and interactive visualisations.Primary reference ↗	Turn numerical work into labelled figures that can be inspected and shared.	Learning context A common way to prepare Python figures. This page displays published assets rather than launching Python plotting sessions.
Machine learning (ML)AI Capability	Methods that estimate patterns or parameters from examples to support inference on other inputs.Primary reference ↗	Learn a pattern from examples, then check whether it helps with new ones.	Learning context ML learns from examples. Read the page-specific methods below to see whether any model runs here.
Deep learningAI Capability	Machine learning using neural networks with multiple representation-learning layers.Primary reference ↗	A layered model learns useful representations; more layers do not guarantee a better answer.	Learning context Deep learning is an AI capability, not a label for every chart. No deep model is implied by this explanation panel.
Natural language processing (NLP)AI Capability	Computational methods for representing, analysing or generating human language.Primary reference ↗	Help a computer work with words, while remembering that context matters.	Learning context Text search, translation, thematic classification and generation are different language tasks; their implementation is identified separately.
Sentiment analysisAI Capability	A language-analysis task that estimates an expressed attitude, often as polarity or sentiment categories.Primary reference ↗	Estimate the tone of words—not the private feelings or personality of their author.	Learning context Sentiment estimates expressed tone. It must not be confused with topic classification or a diagnosis of a person.