The company focuses on providing technological solutions to clients for Human Resource Management. My role has been to innovate their offerings with state-of-the-art technology and advanced analytics.Innovation 1: Developed a world-class Large Language Model-based CV Parser and Analyzer by fine-tuning LLMs using QLoRA technique, implementing 4-bit quantization, and deploying the models on AWS SageMaker. This guaranteed 99% parsing accuracy and saved clients over 100 working hours every month.Innovation 2: Developed and deployed a Machine Learning Pipeline on AWS SageMaker for updating the resume scoring algorithm for each client based on user actions, ensuring maximum relevance of ranked resumes.Innovation 3: Developed and deployed an automated email-based leave approval system as a new product offering for clients, increasing the company's revenue by over 1000 OMR monthly.Innovation 4: Developed and deployed a Human Resource Assistant AI chatbot for generating payslips, checking leave balances, calculating gratuity amounts, detailing benefits/perks offered by the company, etc., saving over 100 working hours monthly and streamlining internal communications for each client.Innovation 5: Automated internal database processes using Python scripts, saving over 700 OMR in monthly employee costs.Innovation 6: Developed over 5 interactive Power BI Human Resource Dashboards highlighting demographic, compensation, recruitment, training, performance, and attrition analytics, and embedded these dashboards in client applications using the Power BI Embedded analytics service, significantly improving decision-making for clients.

Conducted exploratory data analysis (EDA) for CinemaScore prediction and box office performance prediction, producing twocomprehensive reports detailing key insights and findingsDeveloped two machine learning pipelines to forecast CinemaScores and box office performances, achieving high accuracy andproviding detailed performance reportsCreated Excel VBA Macros for forecast generation, integrating Python code and machine learning models, simplifyingthe forecasting process for the client

I focused on developing various statistical and machine learning models for precise sales forecasting. I built an ETL pipeline using SQL, Python, and Apache NiFi, optimizing data storage and processing. I also employed Amazon Redshift for efficient data warehousing to bolster data-driven marketing strategies. Furthermore, I devised a Python-based model for optimizing trade marketing budgets. Employing Agile and Scrum frameworks, I integrated data science models into our sales platform and created interactive dashboards for KPI analysis, streamlining decision-making processes through automated data querying and reporting.

In this role, the focus was on enhancing data management and analysis. This involved consolidating over 20 diverse databases and data sources, like field reports and APIs, into a more cohesive system. A significant part of the job was automating data cleansing and transformation processes using Alteryx, streamlining the workflow. Additionally, there was a strong emphasis on maintaining data integrity, which was achieved by implementing detailed quality checks using SQL and Python, especially for ad-hoc analysis tasks.