Biochemist.
Then self-taught engineer.
Then AI founder.

The first data

Biochemistry degree done. First real job: entering health records for the national HIV programme into a digital system. Over 10,000 paper records, one by one. It was tedious. It was also the first time I understood what it meant to turn chaos into structured, usable data, and why it matters.

Teaching myself the craft

No formal computer science background. Taught myself JavaScript, then frontend engineering, then UX, building enterprise apps for a fintech in Abuja. Cut page load times by 30%. Won design awards. Figured out I could build things people actually used.

Scaling up

Three years at ByteWorks building platforms used nationally: NIMC identity management, MTN SIM registration. Started as a UX designer. Left as a software engineer. The lesson: in a small team, you become whatever the product needs you to be.

First company

Founded Smart Abule Ltd. Built a smart home energy management system from scratch: remote appliance control, real-time metering, automated billing. Collaborated with an embedded systems engineer to wire Arduino hardware to a cloud interface. Working prototype, end to end. Registered. Real.

The drone pipeline

At Keele, I explored how drone imagery and machine learning could support salt marsh species mapping. The v1 research combined SegFormer B5 segmentation with ConvNeXt classification. The manuscript reports mean segmentation IoU of 0.557 and mean F1 of 0.606; its 99.02% classification accuracy is a training-set result, not evidence of performance on unseen imagery. That research became the foundation for EcoVision. I am developing v2's software and evaluation infrastructure, with broader habitat validation and end-to-end evaluation still ahead. Field surveys remain necessary; the research does not establish an automated compliance-reporting service.

Read the EcoVision case study ↗ · Research manuscript (PDF) ↗

Building the architecture

Registered Benched AI Systems Limited and began developing Vaeden (originally StackP), an AI-native engineering assessment platform; CallsAid, multilingual voice AI; and Subra | The AIN Registry, focused on AI agent identity and accountability. These projects deepen the engineering and product experience I bring to a team.

What I'm actually doing

Working as a research assistant, developing EcoVision and Subra, and making time for personal projects. I’m looking for an AI/ML engineering role where I can apply that experience with a team, learn from other engineers, and contribute to shared product goals. My projects give me practical problems to learn from and lessons to bring into that work.