I am a Software Engineer focused on building robust backend architecture, scalable data infrastructure, and efficient distributed systems. I actively contribute to major open-source ecosystems (like Kubeflow) and thrive on optimizing system latency, engineering full-stack applications, and writing production-ready, heavily tested code.

• 🔭 Currently working on: Core architectural improvements and bug fixes for Kubeflow (Katib and Docs-Agent).
• 🌱 Focusing on: Advanced system architecture, scalable data pipelines, and low-latency processing.
• 🤝 Community: Core Technical Member at MATRIX & TechnoMech Clubs, mentoring students and deploying campus-wide tools for 5,000+ users.

Languages:

Frameworks & Libraries:

Tools & Infrastructure:

Python, Spatial/Temporal Frameworks, OpenCV

• Engineered a complex data pipeline combining spatial and temporal video analysis to achieve real-time anomaly detection in CCTV streams.
• Designed a parallel streaming and inference pipeline using Python multiprocessing, ensuring low-latency video processing for continuous surveillance.
• Implemented automated image processing algorithms for activity recognition, significantly reducing the manual effort required for threat identification.

Next.js, Predictive Data Analytics, Geospatial APIs

• Built a data-driven safety navigation system that identifies safer travel routes by analyzing real-time factors like location data, crime rates, and proximity to emergency services using predictive data models.
• Developed real-time tracking and emergency response features, enabling SOS alerts and live location sharing to improve user safety during travel scenarios.
• Designed and integrated a scalable mobile architecture (Android + backend services) with location-based APIs and data pipelines to support continuous route evaluation.

Next.js, TypeScript, Tailwind CSS, Supabase

• Built a production-ready, full-stack hotel booking platform optimizing for fast load times and responsive UX.
• Architected a secure backend using Supabase, implementing SMTP-based email authentication to improve user management and reliability.
• Deployed the application via Vercel with modern CI/CD workflows, ensuring high availability, consistent uptime, and seamless updates.

Python, OpenCV, MediaPipe, NumPy

• Implemented image-based keypoint detection algorithms to classify human poses in real-time, greatly improving posture accuracy detection.
• Mapped dynamic body joint angles to predefined pose templates to deliver automated, instant feedback for user posture correction.
• Optimized the video frame processing loop and integrated lightweight structural models to guarantee efficient processing and a smooth user experience.

• Open Source Architect (Kubeflow): Successfully merged core architecture and bug-fix pull requests into Katib (Automated Tuning) and Docs-Agent. Fixed trainer startup crashes by implementing a robust JSON parser for training parameters with comprehensive pytest coverage (PR #2656, PR #186).
• DSC Winter of Code 2026: Selected from a highly competitive applicant pool as an open-source contributor to maintain and scale community-driven repositories, focusing on bug fixes and automated CI/CD checks.
• #T20DSAChallenge Competitor: Completed an intensive data structures and algorithms challenge, demonstrating advanced proficiency in Stacks, Queues, and complex algorithmic problem-solving.