Ejiaka Nwaeze

I bridge the critical gap between advanced AI models and production-grade software reliability.

My career is built on a unique intersection of clinical science and software engineering. Starting with a foundation as a Medical Laboratory Scientist specializing in Chemical Pathology, I developed an uncompromising eye for data integrity and diagnostic accuracy. Today, I leverage that analytical rigor as a QA Architect and AI Evaluator, specializing in Human Intelligence Software Testing (HIST).

Currently, I orchestrate and audit LLM-driven automated testing pipelines (such as those generated by Claude). Instead of relying on surface-level UI checks, I design test architectures that intercept backend API payloads, validate database state management, and inject "pathological" edge cases to prevent the silent failures that automated AI often misses.

Beyond testing software, I am deeply engaged in evaluating the AI models themselves. I am finalizing my M.Sc. in Bioinformatics, with my thesis focusing on algorithmic fairness and counterfactual recourse in Chronic Kidney Disease (CKD) prediction. This academic background allows me to audit machine learning models not just for code functionality, but for ethical safety, bias mitigation, and clinical validity.

My Technical Edge & Interests:

• AI & LLM Evaluation: I hold an NVIDIA Associate Certification in Generative AI and LLMs, giving me a structural understanding of how these models are built and where they are most likely to break.

• Edge Computing: I actively build and deploy local AI models using edge architecture, specifically the NVIDIA Jetson Orin Nano.

• Quality Philosophy: I believe that AI is a powerful engine, but it requires strict human-in-the-loop oversight to ensure it tests what the business actually needs, rather than just marking its own homework.