- Supervised and mentored a team comprising data scientists and data engineers, fostering a collaborative and high-performing work environment.- Collaborated with the hiring team to identify talent needs and actively participated in the recruitment process to onboard new team members.- Formulated and implemented individualized development plans for team members, providing guidance and support to enhance their skill sets and career growth trajectories.Project : Parking Space Detection- Collaborated with a leading European automobile technology supplier to develop software for parking space detection, overseeing the entire project lifecycle from data requirements definition to model training and integration.- Led the team in various tasks including data quality testing, video parsing, parking space labelling, object detection model training, and object tracking resulting in the successful creation of a robust parking space detection system.- Integrated diverse data sources such as GPS data, camera feeds, and CAN bus data logs to enhance object detection and tracking while also determining the absolute gps coordinates and dimensions of the parking space. Utilized techniques like jit-based compilation and parallel processing to optimize processing speeds.- Played a key role in advancing client technology from TRL 2 to TRL 7 by providing strategic guidance, project management, and technical expertise, demonstrating a significant increase in technology readiness and maturity.- Managed project timelines, contingencies, and resource allocation to ensure smooth execution, maximize efficiency, and minimize risks, leading to timely deliverables and optimal resource utilization.

• Led development of Data Linkage Project to merge data from different sources to create a master dataset using PySpark, S3 and Databricks and created pipelines to link data in subsequent batches leading to an increment from 70% to 95% accurate mappings.• Developed custom string-matching algorithms to match the main titles and utilized ML models including random forests and logistic regression with cross features to match metadata. • Monitored performance metrics of deployed solutions using MLflow detecting data drifts and concept drifts introduced after deployment. • Developed an Emotion Detection Model using fine-tuning and transfer learning on a pre-trained BERT model to detect multiple emotions displayed by users in a tweet directed at client’s twitter handle. • Analyzed video consumption patterns using Markov chains and Dijkstra's algorithm to uncover the most frequented viewing sequences, leading to the development of a recommendation engine.

• Analyzed MaxDiff survey results for an FMCG using Hierarchical Bayesian models to calculate individual and aggregated utility scores. Used Factor Analysis and constructed driver trees to understand consumer preferences across emotional and behavioural latent dimensions.• Implemented an alphabet-level LSTM (Long Short-Term Memory) network to detect fake names in listings, improving data integrity and user verification leading to an 84% reduction in fraudulent listings.

• Developed interactive dashboards using D3.js to help clients analyze television viewing patterns, program schedules, advertisement timings, program reach, ratings etc. Included multiple tabs and filters for enhanced data exploration and insights.• Designed and optimized data tables in MongoDB and DynamoDB including the number of tables, schema, indexing, relationships and partitioning to reduce query latency and costs.

- Contributed to web development projects for a renowned online news platform, utilizing PHP, JavaScript, jQuery, HTML, and CSS.- Introduced enhancements to enhance user experience and functionality on the client's website.

- Safety and Compliance Training- Field visits and Plant Tours- Training and Workshops

Worked on simulating a packed bed heat exchanger and calculated the packing efficiency