RAG - Retrieval Augmented Generation Tool ● Independently developed a prototype tool using an OpenAI LLM to classify and report safety incidents across Intel, processing thousands of reports and categorising incidents into safety themes such as chemical hazards and housekeeping based on predefined criteria. ● Facilitated the identification of major and minor safety concerns through a streamlined AI-powered solution by creating an automated process that enables users to upload incident reports, which are then read by the model, classified into major and minor safety themes, and assessed for report quality (high, medium, low). ● Implemented a three-step process: data ingestion (reading Excel sheets and generating vector embeddings), incident classification using GPT-4o, and output generation of a table with safety themes and quality grading. ● Used LamaIndex for vector embeddings to link user data with the model and Chainlit for building the user interface. ● Deployed additional technologies including MS Azure and SQL to improve safety reporting. Data Driven Risk Tracker Application ● Developed a risk tracker to consolidate data from various departments regarding faulty equipment at Intel facilities, addressing challenges from disparate data formats in individual spreadsheets. Mitigated 70% of high risks associated with the equipment through the tracker. ● Created a unified template for data collection, collaborating with each department to standardise input formats and facilitate accurate data gathering. ● Cleaned and combined the risk data, involving data extraction, transformation, and loading using SQL and Python.Built a risk tracking application with ASP.NET to store and display data, improving accessibility for employees.● Developed a Power BI dashboard to visually represent data, monitor KPIs, and track risk indicators using graphs and charts and employed SharePoint for sharing spreadsheets, uploading guidelines.

● Analysed customer data to identify potential automobile buyers, contributing to business growth through improved targeting. ● Built a machine learning pipeline that involved data preprocessing, customer segmentation using clustering methods, and delivering insights to the marketing team. ● Processed and cleaned customer data, focusing on key demographics to identify profitable customer groups. ● Created a Power BI dashboard to visually represent insights for marketing strategy adjustments. ● Used Python, Sci-kit Learn and SQL to develop models and present findings. ● Collaborated with a team to drive data-informed decisions for client success.

Data Analyst Jun 2018 - Jun 2019 Web Application Developer Jan 2017 - Jun 2018 ● Developed web applications for schools and pharmaceutical companies that previously lacked a unified online platform. Created user interfaces using HTML, CSS, and React.js, ensuring responsive and user-friendly design. ● Built the back-end with Python (Flask) and PHP, integrating the system with a MySQL database for efficient data management. ● Deployed the entire application on AWS, ensuring reliability and scalability for users. ● Implemented JavaScript for dynamic features and interactions across the platform.