Backend, Data & AI Engineer

I build backend, data, and AI systems that hold up in production.

I work across the stack where data meets product: pipelines and data models, services and APIs, and the retrieval and applied AI that sit on top.

5+
Years building
12
Systems shipped
100%
Documented with evidence
01

About

I build fault-tolerant backend systems for data-intensive and AI applications. I design the heavy lifting underneath — event queues, ingestion pipelines, retrieval layers, and the end-to-end observability that ties them together.

The work I enjoy most sits below the surface: moving data reliably from where it lands to where it answers questions, and keeping systems correct under load and graceful when things go wrong. I care about reliability and clarity as much as raw speed.

02

Selected work

Systems built end to end, with evidence.

01

DocuChat

Retrieval-augmented chat over your own documents

  • Ingest pipeline that parses PDFs and Word files, chunks them, and embeds them in batches with retry and progress reporting.
  • Hybrid retrieval over Elasticsearch with citation-linked answers and a grounded refusal path.
  • Correlation IDs propagated end to end — API, queue, and worker — with structured logging for observability.

2026

Sole engineer — architecture, backend, infra

  • Node.js
  • TypeScript
  • Express
  • PostgreSQL
  • Elasticsearch
  • Redis
  • BullMQ
  • Ollama
02

Vertebrae Counter

Detecting vertebrae in knockout-mouse X-rays with a trained YOLOv3 model

  • Trained a YOLOv3 model on the IMPC knockout-mouse X-ray dataset to localize vertebrae.
  • Django API that runs inference and returns bounding-box coordinates for every detected vertebra.
  • React interface that plots detections over the uploaded X-ray with react-bounding-box, containerized for one-command startup.

2025

Sole engineer — model training, API, and interface

  • Python
  • Django
  • YOLOv3
  • Darknet
  • Docker
  • React
03

Toolkit

What I reach for.

Backend

  • Node.js
  • TypeScript
  • Express
  • Prisma
  • PostgreSQL

Data Engineering

  • Python
  • SQL
  • ETL / ELT
  • Data modeling
  • Apache Airflow
  • dbt
  • Apache Spark
  • Kafka

AI & Retrieval

  • Embeddings
  • Elasticsearch
  • Vector search
  • RAG
  • Ollama

Infra & Ops

  • Docker
  • Kubernetes
  • AWS
  • GCP
  • Redis
  • BullMQ
04

Contact

Have a system worth building? Let's talk.

oluwasayoakinkunmi@gmail.com