ibra

Ibrahim AlharthiMachine Learning Engineer

I buildmachinesthat see.

Computer Vision / AI Systems

01 / Projects

Selected work.

01 / Multimodal AI

Roya

Object detection, text recognition and location context in one AI application.

Read case study : Roya
FIG. 01 / SIGNALS INTO CONTEXTCAFETEXT: CAFEVEHICLESTRUCTUREOBSERVED01 / DETECTION + OCRObjects & text02 / GEOSPATIALWhere it isGENERATED03 / LLM INTERPRETATIONAN INTERPRETATION, NOT EVIDENCE
How the pieces connect · illustration, not model output
Read the diagram
  1. Visual input
  2. Detection + OCR
  3. Geospatial context
  4. LLM interpretation
  5. React application
End-to-end platform engineering~1,000 images

02 / Vision-language learning

Learning with less

Adapting CLIP to recognise images with just a few labelled examples.

Read case study : Learning with less
CLIP / SHARED REPRESENTATIONILLUSTRATIVE
Highlight an illustrative image and text concept
“mountain”“flower”“boat”CONCEPTUAL PROJECTION / NOT MEASURED EMBEDDINGS

Images and words. A shared space. Select a concept.

Experiment design · illustration, no measured values
Read the diagram
  1. Image + text inputs
  2. Pretrained CLIP
  3. Few-shot adaptation
  4. Classification evaluation
FIG. 02 / ONE SPACE FOR IMAGES AND WORDSA FEW LABELLED EXAMPLESmountainflowerboatPROMPT“a photo of a”V1V2V3boatIMAGETEXTCLIPPRETRAINEDmountainflowerboatnew image1231TIP-ADAPTER++ CACHE2COOP / COCOOP PROMPT3PEFT ADAPTER
ML experimentationFew-shot

03 / Scientific software

Built for collaboration

Collaborative scientific software, built with an eight-person team.

Read case study : Built for collaboration
FIG. 03 / ONE MODEL, MANY HANDSECOSYSTEM MODEL / SHAREDABCRainfallSoil moistureVegetationGrazingFire riskCarbonHELD BY ABCResearcher AResearcher BSNAPSHOTSv1v2v3nextBACKENDPOSTGISAPI · RBACREALTIMECITESTBUILDDOCKER
23Database tables
14Migrations
32Jest test files
Collaboration sketch · based on the real features
Read the diagram
  1. Shared model editor
  2. Authenticated API + Socket.IO
  3. Per-model permissions
  4. PostgreSQL / PostGIS
  5. CI + tests + Docker
Lead Software Engineer · capstone engagement8-person team

04 / Language models

Inside the language model

Three matched GPT models, with a focus on trustworthy evaluation.

Read case study : Inside the language model
FIG. 04 / SAME RECIPE, THREE RUNSTRAINEVALnever seenBPE / FIT ON TRAIN ONLYk + oo → kookoo + k → kookTOKENSTheĠkookaburraĠlaughsMATCHED HYPERPARAMETERSRUN 0129.9MRUN 0229.9MRUN 0329.9MTOKEN-WEIGHTED PERPLEXITYEVERY TOKEN COUNTS ONCE
Experiment setup · illustration, not a training curve
Read the diagram
  1. Data preparation
  2. Leakage-free BPE
  3. Three matched GPT runs
  4. Token-weighted perplexity
Model training + evaluation3 × 29.9M
More projects
Full-stack commerce Image processing API Web learning map

02 / Engineering experience

Built with teams.
Shipped with care.

Jul 2025 to Jun 2026 / Canberra

Lead Software Engineer

ANU TechLauncher Capstone engagement

Led an eight-person team building a scientific platform for TERN and CSIRO. Owned the backend and CI.

Read the case study
Earlier experience 2022–2024

Feb to Jun 2024 / Riyadh

Software Engineer

Nortal

Built responsive production interfaces in React and JavaScript, working with designers and engineers to turn requirements into features people use.

Jul to Aug 2023 / Taif

Computer Engineering Trainee

King Fahd Air Base

Worked across networking, IP telephony and maintenance: configuring lines, keeping the network running, and tracking down hardware and software faults.

Dec 2022 to Apr 2023 / Remote

Freelance Web Developer

Built and shipped a responsive company website with React and Bootstrap, and looked after its SEO from start to finish.

03 / About

A builder
at heart.

I’m Ibrahim, a machine learning engineer with a background in computer engineering and full-stack development.

Background & education

I trained as a computer engineer, have worked across the whole stack, and spent a year leading an eight-person team through a serious software build. The problems I enjoy most are the stubborn ones, especially when cracking them takes a whole team.

Education

Master of Machine Learning
and Computer Vision

Australian National University · 2024 to present
Expected conferral: December 2026

SDAIA AI Scholarship recipient, 2024(LinkedIn post)

Bachelor of Engineering
in Computer Engineering

Taif University · 2018–2023

Full Stack JavaScript Developer Nanodegree
Udacity · 2022

Skills & tools
Vision & learningPyTorch, CLIP, YOLO, OpenCVAI software systemsPython, FastAPI, ReactEngineering ownershipTypeScript, PostGIS, Docker, CI/CD
Download CV

04 / Contact

Let’s talk.

devibrakh@gmail.com
GitHub LinkedIn Download CV