Open to opportunities

Ikenna Oluigbo

@ikennaoluigbo

AI/ML engineer consulting on agentic AI and RAG systems, with a retrieval workflow that improved candidate and document retrieval efficiency by 35%.

France

Message

As an independent AI Consultant, I advise startups and enterprises on agentic AI architecture, from technical strategy through production deployment. I design reusable AI components and multi-agent retrieval workflows; one improved candidate and document retrieval efficiency by 35% using vector database retrieval.

At INSA Lyon, I architected a ChromaDB-based RAG retrieval system for knowledge-graph tasks in a cross-institutional research environment. I also developed multimodal NLP models that achieved 94%+ accuracy and fine-tuned BERT for domain-specific text generation and sentiment analysis.

At LIRIS, I designed feature-embedding models for multi-domain graph data and proposed a multi-level GNN architecture that outperformed existing models by ~20%. I also built anomaly detection and recommendation models and presented original research at international conferences.

Experience

Work history, roles, and key accomplishments

Current

AI Consultant

Current

Freelance

Nov 2025 - Present (11 months)

Advise startups and enterprises on end-to-end agentic AI system design and architecture, from technical strategy through production deployment. Implemented and deployed frontier LLMs for automation, incident reporting, research, and decision support.

Agentic AI RAG Lang Chain LangGraph MCP Vector Databases LLM Deployment Docker MLOps

AI Research Engineer

INSA Lyon

May 2023 - Sep 2025 (2 years 4 months)

Architected a RAG retrieval system using ChromaDB embeddings for knowledge-graph tasks in a government-funded research environment. Developed multimodal NLP models achieving 94%+ accuracy and fine-tuned BERT for domain-specific tasks.

RAG ChromaDB Docker Streamlit NLP BERT CNN OpenCV YOLO Gan VAES Diffusion Models

Knowledge Graph Research Scientist

LIRIS

Nov 2018 - Nov 2022 (4 years)

Designed novel feature-embedding models for multi-domain graph data using AI and NLP techniques, outperforming prior state-of-the-art approaches. Developed deep learning representation learning models for graph node/edge attributes, achieving 90%+ accuracy.

Knowledge Graphs graph neural networks Deep Learning NLP Anomaly Detection Recommendation Systems ML Pipelines

Machine Learning Engineer

United Nations

May 2020 - Dec 2020 (7 months)

Led development of a real-time data pipeline between a public web portal and PostgreSQL, scaling operational throughput by 40%. Built a deep learning model to predict COVID-19 spread patterns across countries.

PostgreSQL Deep Learning Data Pipelines Plotly Matplotlib Data Analysis

Junior IT Scientist

FUTO

Apr 2013 - Sep 2018 (5 years 5 months)

Conducted research across web and software development domains, managing complex user data and preparing data pipelines for analysis. Designed supervised and semi-supervised ML models for predictions and classifications.

Machine Learning Data Pipelines Web Development Software Development