Building AI systems while advancing model capabilities through research-driven design.
Modular-first solutions across model architecture, inference, and training, with emphasis on scalability, reliability, and performance.
AI Systems Engineering · Inference Infra · LLM & Multimodal AI · Performance Optimisation
- VAAS: Vision-Attention Anomaly Scoring for Image Manipulation Detection in Digital Forensics ·
- Improving Image Embeddings with Colour Features in Indoor Scene Geolocation ·
- Computer Vision for Multimedia Geolocation in Human Trafficking Investigation: A Systematic Literature Review ·
👉 See more on Google Scholar·
👨🏾💻 See Research Engineering Codebase ·
- Evaluation as a System Primitive for Model Behaviour
- Agent-Learning: Execution-Aware Policy Optimization for LLM Tool Systems
- Test-Driven Development in MLOps ·
- ML Systems in Containerized Environments ·
- Production ML Systems on Cloud Platforms ·
More writing: Medium
Learn more: Portfolio