May 2024 – Dec 2024 · 7 mos · Toronto, Ontario, Canada · On-site

• Built a Regulatory Reporting dashboard on PowerBI using PySpark on Databricks, with orchestration and monitoring in Airflow, automating deadline tracking and saving 1+ hours/week; documented on Confluence.
• Automated mortgage balance checks with SQL and Python on Data360, integrating data from APIs and data lake to create a Tableau dashboard that eliminated 3+ hours of manual work and improved accuracy.
• Designed an LLM summarizer to analyze and compare quarterly and annual reports of the Big 5 banks using OpenAI API, leveraging prompts built on Dataiku, enabling summarizations of key insights
• Created an LLM-powered invoice extraction pipeline using Hugging Face models, achieving 95 %+ accuracy in data extraction across 500+ invoice formats, and reducing reconciliation time by 60% for the finance team.

Sep 2022 – Aug 2023 · 11 mos · Delhi, India

Jan 2022 – Sep 2022 · 8 mos

• Built an XGBoost-based time-series forecasting model in Python (Scikit-learn) to predict sales and appointment demand, enabling resource planning and forecasting medicine inventory 3 months in advance; collaborated using Jira.
• Reduced doctor appointment downtime by 85 %+ by analyzing appointment data in Databricks and refining the scheduling algorithm to avoid conflicts.

Jan 2021 – Jan 2022 · 1 yr

• Reduced fraudulent return orders by over 20% by implementing an RFM model to identify and flag exploitative user behavior
• Increased sales by integrating a Naive Bayes machine learning algorithm to generate personalized next-purchase recommendations

Jun 2019 – Jul 2019 · 1 mo · Kuwait

• Worked with the IT division of Kuwait Oil Company, gaining hands-on exposure to day-to-day IT operations, including system support and infrastructure monitoring