Backend engineer. I build systems that hold up when things get real.
Most of what I've shipped is backend systems that can't afford to be wrong — payments, auth, job scheduling, reconciliation. I've also led full-stack teams when the job called for it. Backend is home base, not a boundary.
Languages: Python, JavaScript, TypeScript, Bash, SQL
Core: Django, FastAPI, Django Ninja, Django REST Framework, Next Js, React
Data: PostgreSQL, MongoDB, Redis, MySQL
Infrastructure: Docker, Celery, Nginx, Gunicorn, Uvicorn
Auth: JWT, OAuth, API Keys, Webhooks, RBAC
Tools: Neovim, Git, Postman, Linux
Settle · frontend Payment collection infrastructure for Nigerian SMEs, built on Nomba's virtual account APIs. Every customer gets a dedicated account number so incoming transfers auto-reconcile — no more manually matching who paid what across a shared account. Idempotent webhook processing, an append-only ledger, and two separate SSE channels for real-time payment status. Built solo in under a week.
Flint · frontend A background job scheduler built from scratch — no Celery, no Redis queue. An in-memory heap handles priority dispatch with aging, so low-priority jobs are mathematically guaranteed to eventually run instead of starving forever. A DAG engine handles job dependencies with cascade retry and cascade failure. Dead-letter queue with threshold alerting for jobs that fail for good.
Galleria · frontend Event photo platform using face recognition. Attendees scan their face once and get every photo they appear in, instead of scrolling through hundreds of images. RetinaFace + ArcFace running through Celery workers, embeddings stored in MongoDB, cosine similarity search across thousands of faces in under 100ms.
Insighta Labs · web · CLI Demographic-intelligence API with a dual OAuth flow — standard cookie-based auth for the browser, and a full PKCE flow with a local callback server for the CLI. Short-lived JWTs paired with single-use hashed refresh tokens, and a rule-based natural-language query parser for filtering people by age, gender, and country from plain text.
HNG Internship — Finalist in a highly competitive, global backend internship. Tech lead over an 11-engineer team building an AI-agent tooling product, and built Insighta Labs solo. Before that, built the admin system and notification system for Sitelytics, a website analysis platform.
Taskingdom — Backend developer on a task-based gaming platform with real-money rewards. Built fraud prevention logic, a hybrid Redis/PostgreSQL state tracking system, and webhook integration for third-party game outcomes.
Meerge Africa — Owned the Supplier service from scratch. 500+ suppliers, a four-state approval workflow, JWT service-to-service auth, and 8 weeks as acting backend lead when the engineering lead left mid-project.
Systems that make hard problems boring — priority queues, reconciliation engines, real-time infra. Lately I'm curious about where AI-assisted tooling actually earns its place in a backend engineer's workflow, beyond autocomplete.
oyebowaleabdulmuiz@gmail.com · muizzyranking.me · Twitter
:wq



