Nayim SalamEngineering Portfolio

Systems under
pressure.

I turn messy workflows, signals, and deployments into testable systems with explicit state, guarded decisions, and operational evidence.

BSc Software Engineering · Sail Databank intern · commercial systems operator

View WorkContact

Proof artifact

Edge-AI Seat State Monitor

Noisy seat signals → classified state → guarded runtime mode.

Applied AI prototype

Pipeline flow

01

Signals

Seat & cabin telemetry

pressure / loadseat accelcabin ref
02

Classifier

ML state hypothesis

EMPTYOCCUPIEDVIBRATION
03

Guard logic

Deterministic bounds on model output

τ < 0.65 → holdfault → fallback
04

Bounded mode

Operational output — not direct HW control

monitoringdiagnosticno-action

Operational decision

Low confidence does not trigger action.

It routes to fallback or no-action mode. The classifier informs — guard logic decides what the runtime may emit.

Low-confidence inference → NO_ACTION_UNCERTAIN_LOW_CONFIDENCE

Open case study →

Beyond builds. Where the work is grounded in practice.

Operating context

Education, industry, and commercial systems behind the work.

Education

BSc Software Engineering

Swansea University · 2022–2026

Industry exposure

Software Developer Intern

Sail Databank · Python, testing, documentation

Commercial systems

£50k+ e-commerce venture

Supplier, inventory, logistics, quality, advertising

Full depth. Open a case study for the full decision trail.

Case studies

Explore each build — problem, system shape, and what held.

Every case study covers the problem, actors and state, technical decisions, implementation, limitations, and next iteration.

View WorkContact