Backend Software Engineer. Production-grade APIs, real-time data pipelines, AI-integrated systems — built to run quietly at 3am.
THE ARCHIVE.
ICE
Multi-tenant payment middleware on Nomba's DVA APIs. BullMQ job queues with exponential backoff, auto-refund engine, reconciliation, and a Next.js webhook simulator. Built for Nomba Hackathon 2026.
→ Click to view documentation
Always extending the tech matrix
FEATURED PROJECTS.
EDGE-IQ
AI-powered prediction market trading platform aggregating Polymarket and Bayse data. Generates quant signals via Gemini AI with Kelly Criterion staking, real-time SSE streaming, Celery async workers, and 4-hour conditional caching. Full stack: Django DRF backend + React/Vite frontend.
SURVEYTOOLS KYC
9-milestone KYC backend for a research platform. Multi-layer respondent profiling, AES-256-GCM field-level encryption, NDPR consent and Right-to-Erasure, progressive profiling, and full Jest coverage.
NACOS OAU PLATFORM
Official digital platform for OAU's computing student association. Dues payments, event management with ICS delivery, governance tenure scheduler, Sentry observability, and nightly k6 load tests.
MERIT
University admissions platform with POST UTME score prediction, course recommendation engine based on O-Level and UTME credentials, and an AI chatbot powered by Gemini. Covers 10+ universities.
PEACEMAKER
CLI tool replacing standard git merges with intent-aware, AI-powered conflict resolution. Tier-based classification applies direct merges, AI replays, or generates diagnostic reports automatically.
FACE-ATTEND
Real-time facial recognition attendance system for SWEP 2025 at OAU. Migrated pipeline from LABPH to Dlib, integrated Google Drive API sync, served as Scrum Master, managed Git LFS for large models.
THIS PORTFOLIO
You are here. Hand-built with vanilla HTML, CSS, and JavaScript. Canvas particle background, custom cursor, animated hero, bento grid archive, horizontal project scroll, and a live clock. No frameworks, no bloat.
WORK PRINCIPLE.
Every service, endpoint, and schema must earn its place. No unnecessary abstraction, no premature optimization, and no feature that hasn't been earned by a real problem. Just clean layered architecture, strict type safety, and deliberate engineering decisions.
As a backend engineer working across Python and Node.js ecosystems, I specialize in production-grade API design, async task pipelines, AI-integrated systems, and security-hardened infrastructure. I care about the details that never make it to the README — the ones that keep systems running quietly at 3am.