Experience

AI/ML Software Engineer at EMS Connectivity Solutions, a UK connectivity and AI company based in Hampshire. Working on the AI and machine learning stack across EMS's product line, contributing to the design and engineering of production ML systems that serve real customers. Role reports into the CEO and combines applied ML engineering with research-informed decision making.

Founded and lead Neura-X, a research-grounded AI venture at the intersection of applied AI and enterprise digitalisation. The firm translates frontier AI (RAG, agentic pipelines, efficient transformers, trustworthy ML) into production systems. Now operating in London with plans for expansion.

Curium pioneers Continuous Dynamic Calibration™ (CDC), a proprietary multi-sensor calibration platform spanning LiDAR, Radar, and Camera modalities, enabling rapid escalation of autonomous systems to Level 4 and 5 automation. Developed ML pipelines and AI algorithms for automated multi-sensor calibration across ADAS and autonomous vehicle deployments. Role concluded to focus on new opportunities.

Promoted from Research Associate to AI Researcher and Co-Investigator within months of joining, in recognition of independent research capability and consistent contribution to the lab's publication pipeline (A* conferences and top-tier IEEE/ACM journals). Designing multi-frequency hierarchical learning frameworks that jointly address scalability, client heterogeneity, and catastrophic forgetting in federated settings. Developing a nested federated unlearning scheme enabling selective knowledge removal without full model retraining. Collaborating with researchers ranked in the top 2% globally by citation impact; co-authors span the UK, US, UAE, Pakistan, Saudi Arabia, and Oman.

Contributed to the design and development of Pakistan's first AI-powered Security Operations Centre (SOC) toolset, a strategically critical initiative for deployment across all governmental organizations, designed to autonomously process and triage 25 million+ logs per day. Designed an end-to-end AI-powered SIEM pipeline from large-scale log ingestion through LLM-based threat classification to automated alert routing, replacing manual SOC Level 1 analyst operations at national scale. Recognised as the youngest engineer selected for Pakistan's National AI Research Team.