NATHANIEL SEBASTIAN / COMPUTER SCIENCE & AI
Open to opportunities

Applied AI.
Useful software.

I’m Nathaniel Sebastian, a Computer Science & AI graduate from the University of Leeds. I build AI grounded in source documents, Python APIs and data pipelines, with clear evidence of what works and where the limits are.

INSIDE THE WORK / SYSTEM OVERVIEW
Policy Copilot
Applied AI
STEP 01Policy documents
STEP 02Retrieve & rerank
STEP 03Verify evidence
STEP 04Answer or abstain
PythonSQLFastAPIRAG & evaluation
BASED IN CARDIFF, UK
63
synthetic benchmark questions
Policy Copilot · repository-documented
24
REST endpoints
EventHub · repository-documented
10,150
synthetic transactions
Ledger Data Quality · repository-documented
01 / SELECTED WORK

The work, with the evidence.

Three projects. The problem, my contribution and the decisions behind the result.

POLICY COPILOT / REPOSITORY FIGURERepository workbench figure showing a remote-work policy question, a supported answer and paragraph references.
Workbench figure from the public dissertation repository.
01 /Applied AI

Policy Copilot

Policy questions answered against source documents, with paragraph citations and a clear abstention when the evidence is insufficient.

PythonRAGBM25StreamlitEvaluation

Key decisionMeasure answer usefulness alongside grounding.

Academic dissertation
EVENTHUB / SYSTEM NOTES
GET/events
POST/events/{id}/rsvps
GET/events/{id}/provenance
AUTHENTICATION · OWNERSHIP · PROVENANCE
Conceptual overview
02 /Backend engineering

EventHub

A Python API for events and RSVPs, with authentication, resource ownership and traceable imports from Leeds City Council.

PythonFastAPIPostgreSQLSQLAlchemyPytest

Key decisionKeep imported data traceable to its source.

Academic project
LEDGER DATA QUALITY / SYSTEM NOTES
Checks before conclusions.
CONTROLREVIEW OUTPUT
CompletenessMissing fields
ReconciliationControl totals
DuplicatesCandidates
Conceptual overview
03 /Data engineering

Ledger Data Quality

A reproducible Python and SQL pipeline that checks synthetic financial exports, reconciles totals and produces exception reports.

PythonpandasSQLSQLiteData controls

Key decisionKeep screening signals separate from conclusions.

Personal project · synthetic data
THE REST OF THE REPOSITORIES

Beyond the highlights.

Browse public code, search by keyword and filter by language.

Explore GitHub directory
02 / A LITTLE CONTEXT

Technical curiosity.
Practical follow-through.

I studied Computer Science with Artificial Intelligence at the University of Leeds. My work spans policy-document AI, backend APIs and data-quality systems.

I enjoy the part where a technical idea has to meet a real person's needs: understanding the problem, explaining the trade-offs and making the result useful.

2023–2026 / UNIVERSITY OF LEEDS

BSc Computer Science with AI

Upper Second-Class Honours (2:1)

PAID CLIENT DELIVERY / APR–MAY 2023

Delikatesy Miś

A retailer website delivered from direct outreach through requirements, review demonstrations and client acceptance.

HTML / CSS / JavaScript3 review demonstrations
01

Applied AI, with evidence.

Retrieval, evaluation and citation checks. Clear boundaries around what a system can reliably answer.

02

Data & backend foundations.

Python, SQL, APIs and automated checks. Data flows that are easier to inspect, explain and maintain.

03

Requirements & communication.

Turn a nontechnical brief into a manageable scope, use review demonstrations to gather feedback and keep delivery focused.

04

Interested in useful work.

Applied AI, solutions engineering and technical implementation. Open to London relocation and client travel.

OPEN TO ROLES & PROJECT CONVERSATIONS

Have a role or
project in mind?

Get in touch