AI Engineer

LLMs · RAG / Agents · MLOps

AI Engineer designing, building, and deploying production AI systems across cloud and on-premise environments — from data engineering and model development to workflow orchestration and monitored deployment. Hands-on in LLM fine-tuning (LoRA/PEFT), retrieval-augmented generation with citation-grounded evals, conversational and speech AI, and MLOps with Python, PyTorch, Docker, Kubernetes, AWS, and GCP.

Astronomy PhD, UCT (thesis submitted Jul 2026), bringing large-scale pipeline and statistical-evaluation rigor — 443K×1.8B-record cross-matching with Monte Carlo reliability modelling — to production LLM systems.

Download AI CV GitHub LinkedIn

Experience

Machine Learning Engineer (Consultant, Part-time) — Jan 2025–Present, Cape Town. Design and deliver RAG and fine-tuning (LoRA/PEFT) pipelines for production AI applications, spanning embedding ingestion, vector search, and response generation; build conversational and speech AI including multilingual and low-resource use cases from dataset curation through evaluation; containerize and deploy with Docker and CI/CD (GitHub Actions) across AWS and GCP to monitored cloud-hosted inference.
Machine Learning Engineer (Research Intern, Part-time, Remote) — Jun 2023–Dec 2024, Nabafat AI, US. Built and optimised ML systems for NLP, computer vision, and LLM applications; developed LoRA/PEFT fine-tuning pipelines across foundation models; implemented RAG chatbots over vector databases and embedding pipelines; deployed with Docker, AWS ECR, and CI/CD; improved data quality through large-scale data engineering.
Research Data Scientist / Machine Learning Engineer — Feb 2021–Sep 2024, SAAO, Cape Town. Built ML classification and cross-matching models over 500K+ records and multi-catalogue datasets; designed scalable astronomical data-processing and feature-extraction pipelines; shipped reproducible ML workflows supporting peer-reviewed publications.
Machine Learning Intern — Sep 2020–Jan 2021, Ubenwa Health (Remote, Canada). Built clinical-audio feature-extraction and preprocessing pipelines; developed and evaluated Scikit-learn and XGBoost classifiers; improved data quality through feature engineering and cleaning.

Flagship Builds

Flagship · RAG / Agents

Scientific Literature RAG with Citation-Grounded Evals

End-to-end RAG over scientific literature and catalogues: ETL from external APIs, papers, and structured datasets; embedding ingestion with MinIO/Postgres backing; retrieval over OpenSearch with semantic search; LLM synthesis via Ollama-backed services; containerized with Docker for reproducible deployment.

Stack: Python · APIs · ETL · RAG · MinIO · Postgres · Ollama · OpenSearch · Docker

Speech AI: Whisper Fine-Tuning + Low-Resource Accent Data

Whisper fine-tuning for clinical terminology and radiology reporting, paired with curated Nigerian-English accent audio datasets for low-resource ASR. Covers dataset curation, preprocessing, fine-tuning, and evaluation.

Stack: PyTorch · Hugging Face · Whisper ASR · Python

Weather-Disease MLOps Platform

End-to-end MLOps for weather-sensitive disease prediction: experiment tracking (MLflow), data versioning (DVC), orchestration (Prefect), CI/CD (GitHub Actions), containerized deployment, Terraform-provisioned AWS/GCP infra, and monitoring (Evidently, Grafana).

Stack: Python · MLflow · Prefect · DVC · Terraform · AWS · GCP · Docker · Kubernetes

Supporting proof of scale: MeerKAT × Gaia entity resolution — 443,455 radio sources against a 1.8B-record catalog with Monte Carlo false-positive modelling (94% reliability at 2″, 629 counterparts). Read full analysis.

More builds: Brain MRI tumour classification · ETLFlow (Airflow + Spark) · all projects.

Shipped-With Stack

ML / DL: Python · PyTorch · Scikit-learn · XGBoost · Hugging Face Transformers · Unsloth · NVIDIA NeMo Speech
LLMs & Agents: LangChain / LangGraph · Strands Agents · Agno · Prompt engineering · Embeddings · Vector search (FAISS · Pinecone · OpenSearch)
MLOps & Deploy: Docker · Kubernetes · CI/CD (GitHub Actions) · Model serving · AWS (SageMaker, ECR, Bedrock) · GCP (Vertex AI, BigQuery)
Data: SQL · Spark / Dask · MLflow · Prefect · DVC · Terraform · Evidently · Grafana

Professional Trainings

Machine Learning & MLOps

MLOps Zoomcamp (DataTalks.Club)
Machine Learning Zoomcamp (DataTalks.Club)

Deep Learning & Data Science

Getting Started with Deep Learning (NVIDIA)
Accelerating End-to-End Data Science Workflows (NVIDIA)
Fundamentals of Accelerated Data Science (NVIDIA)

LLMs & Generative AI

Building RAG Agents with LLMs (NVIDIA)
Conversational AI Applications (NVIDIA)
Rapid Application Development with LLMs (NVIDIA)
Introduction to NVIDIA NIM Microservices (NVIDIA)

Education

PhD Astronomy, UCT — Expected 2026. Thesis: Multiwavelength Study of the SARAO MeerKAT Galactic Plane Point Sources.
MSc Physics & Astronomy, 2020, University of Nigeria, Nsukka.
BSc Physics & Astronomy, 2014, University of Nigeria, Nsukka.

Leadership & Community

Arize AI Ambassador; Tavily AI Ambassador.
BRICS / IDIA Data Analytics Trainer (2025 & 2026); Mentor, DARA Big Data Africa School; Mentor, SARAO Open Data for Social Impact Challenge.

Contact

Open to AI Engineer roles (remote / relocation). Cape Town, South Africa. Contact: danoegbo@gmail.com · GitHub · LinkedIn

Download AI CV