Dinesh Sambasivam

Contributed to the architecture, development, and deployment of the scalable in-house product, PTS AI Assistant. This solution empowers technicians to efficiently retrieve Ford-authorized service information through voice and text-based queries, improving diagnostic accuracy and repair efficiency.

Led a transformative project focused on implementing an in-house anomaly detection and alerting platform using Anomaly detection and Time series algorithms. Designed, Developed, and Deployed the platform using Python and SQL to identify and alert upstream data issues promptly, automatically detecting and addressing partial data incidents.

Led the development of a Collateral Model for more accurate Cashflow Forecasts, enabling better decisions in Valuation, Performance Analysis, Risk Analysis, Scenario Planning, and Asset-liability Management. Developed and implemented a robust model utilizing Rep lines data to predict cashflow generation by including loan characteristics and various risk parameters.

Led a team of 3 in the development of a Python-based web application, Gate Keeper Ledger Tool, deployed in production to track and decide spending for each cost center globally. Developed a Python-based automated workflow to digitize capital processes, reducing data collection time from days to minutes.

Individually owned the development of Automated Workflow using Python script, integrating SharePoint, SAP GUI, and HANA view to extract, merge, clean, and transform data. Led a team of 2 in the development and deployment of a Production Process Optimization solution using Python to optimize maintenance operations.

Improved KPI like Containment rate and Transfer rate of a chatbot supporting various workspaces for customers. Developed a Python-based classification model to predict chat transfers or containment based on chat logs, segmenting customers for virtual and live agent assignment.

Worked on a research project to develop a live subtitle system that transcribes user speech during conference calls and displays subtitles in real-time. This system was trained and deployed using the open-source deep-learning framework PaddlePaddle.

Developed a Python-based cognitive system that provides recommendations to improve KPIs by learning interplays of configurations, terrains, traffic patterns, and alarms. Model training was performed using Convolutional Neural Networks with Theano and Keras in Python.

Led a team of 2 to develop a Python-based Chat Bot that automatically generates answers based on user queries using Python packages. This model was trained with location details of employees and meeting rooms to provide automatic replies for navigation queries.

Individually owned the automation process of Text Cleaning, Text Pre-Processing, TF extraction, and TF-IDF extraction for improving SERP. This replaced manual work and provided target keywords for SEO and content optimization.

Developed analytical live dashboards using various technical logics and mathematical formulae for the SEO team and Lead Distribution team. These dashboards enabled better decision-making, strategic planning, and identification of trends and performances.

Developed SQL queries to extract raw data from OLTP systems for Data Mapping, Transformation, and Loading into a Data warehouse for analysis. Created OLAP cubes using SSAS for multi-dimensional analysis and analytical dashboards using Tableau and SpagoBI.

Individually owned the development of a Predictive model using the generalized linear model algorithm to forecast sales by store and department. Implemented techniques like post-forecasting adjustments and low-rank approximation on preprocessed data using R programming.

Grade: 7.5 CGPA

Completed a Master of Technology with a specialization in Data Analytics, achieving a CGPA of 7.5. The curriculum focused on advanced data analysis techniques and their applications.

Grade: 7.65 CGPA

Obtained a Bachelor of Engineering in Computer Science Engineering with a CGPA of 7.65. The program provided a strong foundation in computer science principles and practices.

Grade: 86%

Completed Higher Secondary Certificate with an achievement of 86%. This foundational education prepared for further studies.

Grade: 82%

Achieved Secondary School Leaving Certificate with 82%. This marked the completion of secondary education.