Jul 2023 – Nov 2025 · 2 yrs 4 mos · Bengaluru, Karnataka, India · Hybrid

• AI and Machine Learning Innovation: Developing advanced AI solutions, including dashboards, Machine Learning models, and Generative AI technologies for document understanding (table OCR), document processing, recommendation, and ranking. Working on implementing an attribute extraction pipeline from documents to streamline workflows and improve operational accuracy. Currently building a search engine with hybrid search and image search capabilities to enhance document recommendation and improve user experience.
• System Optimization and Deployment: Leading the transition from monolithic systems to microservices architecture, optimizing GPU utilization and scalability. Developing and maintaining AI deployment strategies on Kubernetes, achieving a 50% reduction in inference and training costs, a 5x latency improvement, and an error rate reduction to under 0.001%.
• Project and Budget Management: Managing project timelines, budgets, and resources to ensure the successful execution of initiatives. Balancing multiple initiatives to meet organizational goals while maintaining high-quality outcomes.
• Domain Expansion: Actively expanding into new domains by transitioning from the lighting construction industry to include electrical and plumbing. Leveraging data-driven insights and innovative solutions to support growth and diversification.
• Leadership and Team Management: Currently leading a team of over 5 data professionals, including divisional leaders, with a focus on fostering collaboration and high performance. Mentoring team members to enhance skills, improve retention, and drive ongoing innovation within the organization. Actively engaged in hiring and expanding the team.
• Stakeholder Collaboration: Collaborating closely with business and product stakeholders to gather requirements and address challenges. Continuously translating stakeholder needs into actionable data strategies to drive growth and profitability through innovative AI and data solutions.

Apr 2022 – Jul 2023 · 1 yr 3 mos · Remote

• Working on a projects involving the development of long-term weather forecast, price forecast, and demand forecast models for the Australian Energy market. I implemented an end-to-end solution for electricity generation profiles of all the large-scale grids in Australia.

Dec 2020 – Apr 2022 · 1 yr 4 mos

• > Personalization initiative - Reranking products to provide personalized experience using propensity modelling. I developed a customer purchase likelihood model by analyzing 5 million+ customer transactional data to predict their affinity to specific brands and taxonomy. Through this model, I was able to personalize product displays by re-ranking products on different pages, resulting in a projected additional gain of over 100k$.
• > I mentored a 10-member team to understand customer feedback using social media sentiment analysis tools. This experience allowed me to not only make a meaningful impact in improving customer experiences but also to help my team develop valuable skills in data analysis and customer feedback analysis.

Nov 2018 – Dec 2020 · 2 yrs 1 mo

• Self-service refund initiative - Owned the implementation and deployment of machine learning models to automate the refund process. Helped in reducing the refund time from a day to 30 min or less. I developed an end-to-end predictive model to identify fraudulent refund requests for a self-service refund system. Through my work, I was able to automate the handling of 72% of refund requests and provide insightful metrics to customer care agents for informed decision making.
• > Auto-basket initiative - I designed an algorithm to compute personalized consumption cycles of grocery products and developed an automated grocery list that increased shopping efficiency up to 300% less time. These efforts resulted in higher repeat orders and increased revenue of over 100k$. Customers could finish their shopping journey in less than 5 min.

Aug 2018 – Oct 2018 · 2 mos · Bengaluru, Karnataka, India

• Worked on differentiating between normal and abnormal chest X-ray images using Convolutional Neural Networks (CNN).

Sep 2016 – Jul 2018 · 1 yr 10 mos · Bengaluru, Karnataka, India

• I developed a fashion recommender system for retail stores. We utilized object detection to identify clothing items from images and implemented an image-based recommender system to suggest similar items from the store's catalogue. Additionally, we developed a novel multi-task multi-class model using a deep learning approach to predict consumers' race, gender, and age for consumer analytics. Our model achieved 95%+ accuracy on gender and 75%+ accuracy on age, surpassing existing algorithms. We also reduced the total number of parameters by 10x, allowing the model to run on CPU-only ARM devices and increased the inference speed by 4x compared to our existing implementation.

Dec 2015 – Sep 2016 · 9 mos

• I helped 42 clients in developing statistical and machine learning based models to make the most out of their data. Some of my key projects are -
• ✔ Applied Decision Tree and Multivariate Regression algorithms to predict industrial combustion gas emissions to optimise operational conditions
• ✔ Implemented the Artificial Neural Network to forecast future sales for rationalising inventory
• ✔ Conducted outlier detection on soccer data and applied the Support Vector Machine Learning
• algorithm to predict match outcome with an accuracy above 65% (chance 33%) for bidding.

Jul 2012 – Aug 2012 · 1 mo · Lausanne Area, Switzerland

• I compared QGel’s product with competitor’s product and identified potential advantages & challenges which led to better decision-making in product launching and advertisement.

Jun 2009 – Aug 2009 · 2 mos · Orsay paris-sud

• Computational modelling of protein motifs to identify potential drugs for further pharmaceutical research.