I build machine learning models, analytics and GenAI applications that commercial teams use to make decisions. I have more than four years' experience across retail, energy, insurance and restaurant operations, first at Infosys and now at Tahir Group.
I work across the full cycle: getting the data right, choosing and testing the model, and presenting the result so the business can use it.
Regression, classification, clustering and time-series forecasting, with the validation needed to trust them in production.
Retrieval-augmented generation, LLM-as-a-judge ranking and multi-model pipelines, built with evaluation and guardrails.
Marketing mix modelling, promotion optimisation, RFM segmentation and difference-in-differences for measuring impact.
Azure Databricks, Data Factory and Azure ML, with Git and CI/CD taking work from notebook to scheduled pipeline.
A multi-LLM job-search agent I use every day, a retrieval-augmented analytics agent and an end-to-end insurance analytics platform.
Case studyA multi-LLM system that collects 2,000+ live roles a day from 9 ATS APIs and job boards, scores each one against my CV with a three-model LLM-as-a-judge cascade, and assists with applications through a Chrome extension that learns from my edits.
MScQuestion answering over business reports, with every answer traced to its source page. Reports in PDF, Word, Markdown or text are chunked, embedded into a FAISS index and retrieved for a Gemini model, with automated answer evaluation and unit tests across the pipeline.
End to endAn end-to-end platform on PySpark and Delta Lake: bronze, silver and gold layers, star-schema marts, and churn, fraud and high-cost-claim classifiers with MLflow experiment tracking, batch scoring, data-quality monitoring, Docker and CI/CD in GitHub Actions.
I'm open to data science, machine learning and GenAI roles. The quickest way to reach me is email.