Sumit Singhal Email & Phone Number

Bengaluru, Ka, In

Bengaluru, Ka, In

Bengaluru, Karnataka, India

Bengaluru, Karnataka, India

I successfully led key projects in Computer Vision, achieving over ₹1M in monthly savings through a custom user onboarding pipeline and improving face and eye detection with Google ML Kit. In Deep Learning, I developed a robust liveliness detection model and optimized hyperparameters with precision. My expertise extended to Generative AI, where I created realistic avatars and implemented advanced models for QR code generation utilizing Stable Diffusion Models and Control Nets. In… Show more I successfully led key projects in Computer Vision, achieving over ₹1M in monthly savings through a custom user onboarding pipeline and improving face and eye detection with Google ML Kit. In Deep Learning, I developed a robust liveliness detection model and optimized hyperparameters with precision. My expertise extended to Generative AI, where I created realistic avatars and implemented advanced models for QR code generation utilizing Stable Diffusion Models and Control Nets. In Algorithm-design, I developed an even user-agent distribution system and improved collections by 24% through language optimization. Data Engineering efforts included pipeline enhancement and RESTful API architecture. Additionally, my work in Natural Language Processing to create customer support AI led to enhanced chatbot resilience and a hackathon-winning chatbot creation. Show less

Kanpur, Uttar Pradesh, India

As the sole first author of a 40-page research paper currently under review at Transportation Research Part C having impressive H-index of 164, I've showcased strong research and analytical skills. I developed a Bayesian ML framework with multiple loss functions that saved 1630 minutes, reducing delays for 83% of flights. I extracted crucial parameters from government datasets and identified 12 key factors affecting flight delays utilizing SHAP plots and WMA charts. Statistical analysis and… Show more As the sole first author of a 40-page research paper currently under review at Transportation Research Part C having impressive H-index of 164, I've showcased strong research and analytical skills. I developed a Bayesian ML framework with multiple loss functions that saved 1630 minutes, reducing delays for 83% of flights. I extracted crucial parameters from government datasets and identified 12 key factors affecting flight delays utilizing SHAP plots and WMA charts. Statistical analysis and machine learning techniques like Ridge regression, XG-Boost, ANN, and LightGBM were used to create a delay prediction pipeline with an impressive 84% F1 score. Furthermore, I leveraged Bayesian optimization for optimizing flight departure times to minimize journey delays. This research demonstrates a profound understanding of the subject matter and expertise in data analysis and machine learning. Show less

Munger, Bihar, India

During my tenure as a Techno Managerial Intern at ITC Limited, I delved into the realm of Industry 4.0, focusing on Real-Time Condition-Based Monitoring for machine KPIs, all under the banner of Six Sigma excellence.My contributions included the identification of 9 crucial factors for machine breakdown, leading to the development of alert conditions using EMWA control charts. This was further augmented by a rigorous Root Cause Analysis conducted within a Six Sigma framework, resulting… Show more During my tenure as a Techno Managerial Intern at ITC Limited, I delved into the realm of Industry 4.0, focusing on Real-Time Condition-Based Monitoring for machine KPIs, all under the banner of Six Sigma excellence.My contributions included the identification of 9 crucial factors for machine breakdown, leading to the development of alert conditions using EMWA control charts. This was further augmented by a rigorous Root Cause Analysis conducted within a Six Sigma framework, resulting in the identification of 10 critical attributes that had been leading to false rejections.The results of this endeavor were remarkable: a 21% increase in production, made possible by the implementation of a monitoring system across the machine's three KPIs through the utilization of Power BI and Python. Simultaneously, there was a 60% reduction in the overall rejection rate within the factory, accompanied by the successful execution of two major projects instead of one as originally planned. Show less