Hey!You can call me Abdur
ML/AI Developer with 6+ end-to-end machine learning projects spanning anomaly detection, forecasting, and NLP — delivering 90%+ model performance and CI/CD-backed deployments. I also do photography on the side for fun c:
Background
I'm an ML/AI developer with a strong focus on building end-to-end machine learning systems — from data pipelines and model training to containerized deployments with CI/CD.
My experience spans anomaly detection, time-series forecasting, NLP, and computer vision. I enjoy turning complex problems into clean, production-ready solutions that deliver measurable impact.
I'm passionate about the intersection of AI and software engineering — building systems that are not only intelligent but also robust, scalable, and maintainable.
I also like doing art.
(I also like coffee more than chai.)
Education
Bachelor of Computer Science
Majoring in Big Data & Artificial Intelligence
University of Wollongong in Dubai
Relevant Coursework
Core Competencies
Programming & Data
ML & AI
Deployment & MLOps
Computer Vision
Prooooooooojects
End-to-end machine learning and full-stack projects — from research to production-ready deployments — These are some of the projects I've worked on.
Neurocode turns GitHub repositories into living documentation and structured onboarding, cutting new-developer ramp-up from weeks to hours. It provides code-aware chat, task-scoped guidance, and change-risk detection using retrieval-augmented generation. Interactive architecture maps and auto-generated UML diagrams help engineers understand large codebases in seconds.
Neurocode
Unsupervised defect/anomaly detection for MVTec AD that outputs reconstruction-based anomaly heatmaps and segmentation masks. Designed for fast inference so defects can be localized and acted on quickly.
Defect Detection
Predicts parcel delivery surge levels from delivery history and public holiday signals to automate planning. Includes forecasting to project upcoming surge patterns.
Surge Classifier
Automates the discovery and formatting of citation-worthy sentences, improving how quickly relevant references are identified. Produces context-aware re-citations to support consistent citation workflows.
Citera
Helps developers discover and showcase GitHub repositories with AI-generated feedback and documentation derived from repository structure. Streamlines repository understanding so projects can be shared faster.
Sharded
Neurocode
Neurocode turns GitHub repositories into living documentation and structured onboarding, cutting new-developer ramp-up from weeks to hours. It provides code-aware chat, task-scoped guidance, and change-risk detection using retrieval-augmented generation. Interactive architecture maps and auto-generated UML diagrams help engineers understand large codebases in seconds.
Defect Detection
Unsupervised defect/anomaly detection for MVTec AD that outputs reconstruction-based anomaly heatmaps and segmentation masks. Designed for fast inference so defects can be localized and acted on quickly.
Surge Classifier
Predicts parcel delivery surge levels from delivery history and public holiday signals to automate planning. Includes forecasting to project upcoming surge patterns.
Citera
Automates the discovery and formatting of citation-worthy sentences, improving how quickly relevant references are identified. Produces context-aware re-citations to support consistent citation workflows.
Sharded
Helps developers discover and showcase GitHub repositories with AI-generated feedback and documentation derived from repository structure. Streamlines repository understanding so projects can be shared faster.