Daniel Egbo

Machine Learning & AI Engineer

Building LLM applications, RAG systems, and ML pipelines. PhD researcher applying the same scientific rigor to large-scale radio-astronomy data at UCT/SAAO.

🎯 Target: ML Engineer / AI Engineer roles in applied ML or LLM infrastructure

Cape Town, South Africa (Open to Remote / Relocation)

Google Scholar | LinkedIn | GitHub | Twitter

Download CV GitHub

6+

Publications

15+

Projects

4

Awards & Grants

5+

Certifications

About Me

I am an Astronomy PhD Candidate at the University of Cape Town and the South African Astronomical Observatory, researching active radio-emitting stars using multi-wavelength data from MeerKAT, Gaia, and eROSITA. My work involves cross-matching millions of astronomical sources, optical spectroscopy, and statistical analysis to understand stellar magnetic activity.

Alongside my research, I build machine learning systems, LLM applications, and data engineering pipelines. I have developed RAG-based chatbots, computer vision classifiers, fraud detection models, and fine-tuned large language models. I combine scientific rigor with practical engineering to solve real-world problems with AI.

I am actively seeking industry opportunities where I can apply my expertise in machine learning, data science, AI engineering, and scientific computing.

Featured Projects

⭐ Featured Project (GenAI / Agents)

Multi-agent RAG Applications

Developed a multi-agent retrieval-augmented generation (RAG) system utilizing specialized LangGraph agents to orchestrate document-based question answering.

Impact: Achieved a 30% reduction in hallucination rates and improved response accuracy by 25% on complex scientific papers.
Features: Automated routing, query reformulation, self-correction, and citation verification mechanisms.

Tech Stack: LangGraph, LangChain, LiteLLM, Qdrant, Milvus, Python

Medical Speech Recognition using Whisper

Fine-tuned OpenAI's Whisper model on clinical dictation datasets to transcribe specialized medical nomenclature, radiology reports, and accented audio.

Impact: Reduced Word Error Rate (WER) by 15% for specialized terminology over out-of-the-box baselines.
Scale: Processed and optimized workflows for 200+ hours of medical recordings.

Tech Stack: PyTorch, Hugging Face, Whisper ASR, Python

Stellar Counterpart Identification in the MeerKAT Galactic Plane Survey

Cross-matched 443,000+ radio sources from the SARAO MeerKAT survey against Gaia DR3's 1.8 billion-object catalog, using Monte Carlo simulation to statistically quantify match reliability and rule out chance alignments. Identified 629 candidate stellar counterparts.

Impact: The largest radio-optical crossmatch sample of its kind.
Scale: Cross-matched 443,000+ radio sources from the SARAO MeerKAT survey against Gaia DR3's 1.8 billion-object catalog.

Tech Stack: Astropy, TAP, TOPCAT, Monte Carlo simulation

View All Projects Data Science Portfolio

Technical Skills

ML & AI Engineering

PyTorch · Scikit-learn · XGBoost/LightGBM · Hugging Face · LangChain/LangGraph · RAG systems (Qdrant, Pinecone, Milvus) · Fine-tuning (Unsloth) · Agentic systems · OpenAI/LiteLLM APIs

Data Engineering & Cloud

Python · SQL · Pandas/NumPy · Airflow/Prefect/Kestra · dbt · BigQuery/DuckDB/Postgres · AWS/GCP · S3/GCS/MinIO

Scientific Computing (PhD research)

Astropy · TOPCAT · Large-scale catalog cross-matching · Statistical/Monte Carlo methods · NVIDIA RAPIDS (cuDF/cuML)

Selected Publications

Egbo, O. D., Groot, P. J., Buckley, D. A. H., Robrade, J., Schwope, A. D., Freund, S., Schneider, P. C. & Stelzer, B. (2026). X-ray counterparts to stellar MeerKAT Galactic-plane compact radio sources Astronomy & Astrophysics, 707, A193.
Egbo, O. D., Buckley, D. A. H., Groot, P. J., Cavallaro, F., Woudt, P. A., Thompson, M. A., Mutale, M. & Bietenholz, M. (2025). The stellar population in the SARAO MeerKAT Galactic Plane Survey Monthly Notices of the Royal Astronomical Society, 540(3), pp.2685-2702.
Potter, S. B., Buckley, D. A., Scaringi, S., Monageng, I. M., Egbo, O. D., Charles, P. A., ... & Hlakola, M. (2024). Optical spectroscopic and photometric classification of the X-ray transient EP240309a as an intermediate polar Monthly Notices of the Royal Astronomical Society: Letters, 532(1), L21-L26.
Garnavich, P., Potter, S. B., Buckley, D. A., van Dyk, A., Egbo, O. D., Littlefield, C., & Greiveldinger, A. (2023). Rapid Evolution of the White Dwarf Pulsar AR Scorpii The Astrophysical Journal Letters, 958(2), L22.

View all publications & presentations

Honors and Awards

SAAO PhD Prize Scholarship Award for PhD study in Astronomy at the University of Cape Town, 2021-2024
Breakthrough Listen Travel Grant to attend the 2023 ThunderKAT meeting at the University of Oxford, September 2023
COSPAR Grant to attend the X-VISION School, a Joint IAU I-HOW and COSPAR Capacity Building workshop at North-West University, South Africa, February 2023
AGNES Intra-Africa Mobility Grant to visit SAAO for research during MSc, 2018

Professional Training & Certifications

NVIDIA Deep Learning Institute

Building Conversational AI Applications (2025)
Accelerating End-to-End Data Science Workflows (2024)
Getting Started with Deep Learning (2024)
Generative AI with Diffusion Models (2024)
Building RAG Agents with LLMs (2024)
Disaster Risk Monitoring Using Satellite Imagery (2024)

Data Science & ML

MLOps Zoomcamp - DataTalks.Club (2025)
Machine Learning Zoomcamp - DataTalks.Club (2023)
Applied Data Science II: ML & Statistical Analysis - WorldQuant University (2023)
Applied Data Science I: Scientific Computing & Python - WorldQuant University (2023)

Summer Programs

Oxford Machine Learning Summer School (2023)
COSPAR X-Vision School: X-ray Astronomy (2023)
ESCAPE Summer School: Data Science for Astronomy (2021)
ZTF Summer School: Time-domain Astronomy (2021)
GROWTH Astronomy School: Time-domain Astronomy (2020)