Adejare Adelugba
// the author abuja, 2026
[ WHO I AM ]

About.

I'm a computer-science graduate from Landmark University who does machine learning where the data is worst — hospital imaging archives, half-scanned government records, spreadsheets nobody trusts. My thesis fused MedCLIP image embeddings with TabNet clinical features and beat both single-modality baselines. Two internships taught me that shipped automation beats clever prototypes. Every project I publish documents what didn't work, not just what did.

[ CAREER ]

Experience.

// 2021 — present
  • automated the firm's weekly consulting reports end-to-end: data collection to formatted memo
  • designed, built, and shipped the corporate website
  • automated document-processing workflows across departments, cutting manual handling time by ~90%
  • built validation scripts that caught entry errors before they reached official records
  • thesis: multimodal breast-cancer detection — image + clinical fusion, beating unimodal baselines
  • focus areas: machine learning, databases, distributed systems
[ TECHNICAL MATRIX ]

Capabilities & Stack.

// verified tools & frameworks
// ml-ai
pytorch scikit-learn medclip tabnet dinov2 grad-cam
// data-engineering
postgresql selenium pandas etl pipelines schema validation
// automation-web
python fastapi docker html5/css3/js
SYSTEM_ARCHITECTURE
├── Multimodal ML // Vision (MedCLIP/ViT) + Tabular (TabNet)
├── Resilient ETL // Selenium scraper pool + PostgreSQL validation
└── Deployment // Docker containers + Cloudflare static edge hosting
[ QUICK FACTS ]

At a glance.

// for the 90-second reader
education
B.Sc. Computer Science, Landmark University
focus
Multimodal ML · medical imaging · data infrastructure
stack
Python · PyTorch · PostgreSQL · FastAPI · Docker
location
Abuja, Nigeria — remote-friendly (WAT)
status
Open to ML Engineer & Data Scientist roles

// certifications — deeplearning.ai deep learning specialization · google data analytics · full list on linkedin ↗