Argusa AI Challenge 2025 · Runner-Up 🥈
Secured 2nd place at the Argusa AI Challenge 2025 by building ARGRAG, a RAG system for complex enterprise document corpora with multi-modal analysis, metadata intelligence, and real-time performance.
AI engineer and full-stack builder.
LLM agents, RAG systems, and production ML.
Software Engineer at CERN, shipping AI & HPC tooling for the CMS experiment.
Selected work
Experience
Software Engineer — CERN, CMS Offline & Computing
Developing WMCore and Unified for CMS Monte Carlo simulations and reconstruction across the WLCG. Built React + FastAPI ops dashboards, automated deployments with Kubernetes/ArgoCD, and manage Oracle, Mongo, and MySQL DBs with OpenSearch/Grafana observability.
Research Engineer (CV & ML) — Retrocausal
Shipped SOTA activity-recognition and action-segmentation models into a PyQt desktop client with a Django + VueJS stack on AWS, improving factory-floor accuracy metrics by 10%.
OpenLab Summer Student — CERN BioDynaMo
Automated CI/CD with Docker + GitHub Actions and refreshed the publishing pipeline from Binder to JupyterLab for the BioDynaMo agent-based simulation platform.
BS Computer Science — FAST-NUCES
Graduated with a 3.78 CGPA. Rector and Dean's List honours, focusing on distributed systems, ML, and large-scale web engineering.
Publications — ECCV 2024 · ISMAR 2022 · BioDynaMo 2021
Co-authored work on aligning 2D skeleton sequences with multi-modality fusion, AR job-status tracking, and BioDynaMo lab automation.
Awards & hackathons
2024 CMS Award for modernising Unified operations. Runner-up, Argusa AI Challenge 2025 at EPFL. 2nd prize, GovTech Hackathon 2021 (World Bank + Code for Pakistan). Finalist, CERN Webfest 2021.