Resume

B.Tech in Computer Science & Engineering

• Specialization: Artificial Intelligence & Machine Learning
• CGPA: 8.92/10
• Achievement: Best final year project award, VIT University 2024, for developing an on-premise AI learning platform

Machine Learning Engineer

• Pioneering the development of a neural network-based credit risk model from unstructured SMS data ensuring transparent and interpretable underwriting decisions aligned with regulatory standards
• Accelerated feature generation by integrating a Ray-based head–worker system and optimizing request throughput. Improved feature service performance by tuning thread pools, scaling workers, and refining SQL queries, resulting in an 8× decrease in feature fetch time
• Achieved a minimum of 2% delinquency rate by leading the development of a Credit Risk Model for premium segment users
• Reduced model training time and costs by 5x by building a fully automated pipeline using AWS SageMaker
• Ensured 99.5% data consistency through rigorous QA & QC testing of credit bureau data in multiple formats
• Refined training sample queries and introduced shorter window targets, enabling the use of recent data and improving data confidence & generalizability across all user segments

Machine Learning Engineer

• Developed cashflow and balance classification & extraction SMS model with AUC of 1.00, increasing coverage by 3x
• Engineered product-specific features from credit bureau data, resulting in a 6% improvement in model AUC
• Optimized processing of 100K+ files/min by building a parallel processing algorithm to convert historical data into Credit Information Reports

Research Intern

• Built a parallel and distributed computing workspace using Ray to accelerate ML training and inference
• Configured Linux & ML Frameworks on IBM PowerPC ppc64le architecture
• Architected and deployed an AI Learning Platform on distributed clusters to support hands-on learning for junior students

Machine Learning Engineer

• Developed a real-time Face Detection & Recognition System using YOLOv7 & ArcFace
• Implemented the AI service on DeepStream SDK, with integration of Azure IoT Hub and Qdrant

AI Learning Platform

• Role: Individual Contributor | Team Size: 1
• Built an AI-Learning-Platform, an LLM-RAG pipeline that acts as a guide and resolves students' doubts
• Utilized vLLM for model inference and integrated Qdrant with Langchain for the RAG pipeline
• Developed a backend server using Django, storing user data in PostgreSQL, chat, and feedback in Cassandra

AI Malware System

• Role: Project Lead | Team Size: 4
• Developed a Cross-Platform Application that uploads files to an AI-driven backend for identifying malware with an Azure PowerBI dashboard to visualize the insights in real-time
• Employed CoAtNet Transformer model pre-trained on 2 million malware binaries, and conducted advanced feature analysis using LSTM & LightGBM based on logs from Cuckoo Sandbox and malware header information

Crime Activity Detection System

• Role: Project Lead | Team Size: 4
• Designed a real-time Surveillance AI system using a pre-trained SwinTransformer as an AutoEncoder for spatial feature extraction, integrated with LSTM decoders to model temporal anomalies in video streams from the UCF-Crime dataset
• Enhanced anomaly classification by combining SVM-based scoring with sequence-aware reconstruction loss to capture subtle spatiotemporal deviations
• Built and deployed a WebSocket-based interface for seamless real-time video ingestion and anomaly alerting, architected for edge deployment and multi-camera scalability

• Machine Learning in Production, Coursera – offered by DeepLearning.AI

• AWS Certified Developer – Associate, Udemy

• Runner, Branch Hackathon – For Building a GenAI Credit Risk Model

• Winner, Best Final Year Project, VIT University, for developing an on-premise AI learning platform

• Winner, Branch Hackathon – For developing an AI Agent to assist engineers with documentation

• Top 1% in 10th & 12th Grade Board Exams

• 98.92%ile in Jee Mains, conducted all over India

• Represented Inter-Club Cricket Team as a Left-Arm Fast Bowler & Captain

Project Lead, Machine Learning Initiatives

• Directed end-to-end ML projects including model development, deployment, and performance optimization in production, impacting credit scoring decisions for over 100,000 users

Public Relations Officer

• Spearheaded communication, outreach, and publicity efforts for the club
• Organized fundraising events, coordinated donation drives for underprivileged communities, and managed promotional campaigns to increase club visibility
• Played a key role in executing events such as blood donation camps, cleanliness drives, and educational workshops