As a Senior Full-Stack Developer at Applied AGI, I work across the stack, building high-performance web platforms, optimizing backend systems, and occasionally integrating machine learning workflows. I’m especially drawn to algorithms and data structures, which fuel my interest in writing efficient, scalable code.
Beyond daily development, I enjoy exploring new technologies and continuously refining my skills. I’ve worked extensively with data-heavy systems, modern web frameworks, and infrastructure tools that support real-time applications and robust deployments. What drives me is the challenge of solving complex problems and building tools that create meaningful, measurable impact.
Experience
Senior FullStack Developer | Mar 2023 - Current
Applied AGI, Brentford, London
- Real-Time Video Prediction System: Developed a browser-based interface using React, TypeScript, and WebSockets, enabling dynamic bounding box customization, visibility toggling, and color adjustments for live ML predictions.
- Live Video Streaming Backend: Built a server-side solution for real-time predictions on webcam/video feeds using Django, OpenCV, and PyTorch, implementing efficient chunk-based HLS (m3u8) streaming.
- ML Model Comparison Platform: Created a Next.js application with MySQL and Tailwind CSS for side-by-side model benchmarking, enabling performance evaluation across diverse datasets.
- Advanced Annotation Tool: Engineered a high-performance labeling tool (10k+ lines of TypeScript) supporting bounding boxes, polygons, zoom, and real-time class/annotation management with optimized matrix calculations for smooth interactions.
- Model Performance Analytics: Designed a scalable architecture for statistical evaluation of ML models, enabling real-time comparison and accuracy metrics visualization.
- Infrastructure & Reliability: Managed system services (systemd) for critical applications, ensuring high availability via automated logging, uptime monitoring, and rapid recovery protocols.
- DevOps & CI/CD: Streamlined workflows using Bun.js, PNPM, Conda, and Docker; maintained CI/CD pipelines for issue tracking, PR reviews, and sprint management.
- UI/UX for ML Tools: Delivered intuitive interfaces for annotation and prediction visualization, ensuring seamless user customization and interactivity.
- Monitoring & Testing: Self-hosted Sentry for cross-platform (Next.js, Django, Python) error monitoring; wrote Jest tests and managed APIs via Postman for QA.
- Geospatial Data Integration: Developed export modules for QGIS and ATAK applications, handling large-scale high/low-res imagery and annotations with efficient data workflows.
- System Administration: Automated service deployments, managed containerized environments (Docker), and optimized server resources with Ngrok for secure internal access.
Python Developer | Nov 2022 - Feb 2023
ETL Systems, Rickmansworth, UK
- Full-Stack Development & Data Systems:Developed statistical models and software applications while engineering scalable data pipelines (Python/SQL/S3). Led SQL-to-Cassandra migration (40% faster queries) and created interactive Vue.js dashboards with Plotly/mpld3 visualization.
- Process Automation & DevOps: Automated UNIX workflows (60% time savings) and implemented CI/CD with GitHub Actions/Docker. Managed task allocation, version control, and deployments across staging/production environments.
- Machine Learning & Leadership: Designed CNN architectures (81% accuracy) with OpenCV integration for medical imaging. Directed agile teams in delivering complex biomedical data solutions on schedule.
Software Engineer | Aug 2020 - Sep 2021
Indian Institute of Technology, Indore, India
- Statistical Computing & Software Engineering: Performed advanced statistical analysis and developed optimized applications with focus on computational efficiency and system reliability.
- Medical Data Infrastructure: Architected scalable ETL pipelines for heterogeneous medical datasets using Python, SQL, and AWS S3, implementing data validation protocols for 99.9% ingestion accuracy.
- Database Modernization: Directed the migration from SQL to Cassandra NoSQL systems, achieving 3x throughput improvement through Python automation scripts and Dockerized testing frameworks.
- Clinical Data Visualization: Engineered interactive dashboards using Vue.js and Bootstrap with Plotly/mpld3 integration, enabling real-time exploration of complex biomedical datasets.
- UNIX Systems Optimization: Automated 20+ manual workflows through Bash scripting and Linux utilities, reducing processing time by 70% across research computing clusters.
- DevOps Implementation: Established CI/CD pipelines using GitHub Actions and Docker containers, standardizing deployment workflows across development, staging, and production environments.
- Medical Imaging AI: Developed CNN-based diagnostic models achieving 81% classification accuracy through OpenCV-powered image preprocessing and neural architecture optimization.
- Technical Project Leadership: Coordinated cross-functional teams of 5+ researchers and developers, implementing agile methodologies to deliver complex bioinformatics solutions on schedule.
Fullstack Developer | Jun 2019 - Jun 2020
Raja Ramanna Centre for Advanced Technology, Indore, India
- Real-Time Data Processing: Designed and implemented a scalable data pipeline using PySpark and SQLAlchemy to process semi-structured data from multiple sources, improving data throughput by 40% and enabling real-time analytics.
- Legacy System Modernization: Spearheaded the migration from legacy infrastructure to a modern tech stack (PHP, HTML5, JavaScript, Firebase, SQL Server 2012), enhancing system performance and user experience while reducing maintenance costs by 30%.
- System Enhancement & Visualization: Redesigned core systems with custom role-based access control, interactive dashboards using AmCharts, and integration of mathematical tools (MPS/QPSH), increasing operational efficiency by 25%.
- Full-Stack Development: Built high-performance web modules using Python and Django framework, optimizing both front-end responsiveness and back-end processing speed by 35% through code refactoring and caching strategies.
- AI-Powered Automation: Developed a genetic algorithm-based scheduling system for airline operations, reducing manual scheduling effort by 75% and computational overhead by 60% while maintaining 98% schedule accuracy.
- DevOps & Database Optimization: Deployed and maintained web applications on Linux servers, implemented CI/CD workflows via GitHub, and enhanced SQL Server 2012 performance by 50% through optimized stored procedures and indexing strategies.