Work

Built and led world-class engineering teams across Platform, Compute, and Databases, scaling total headcount to 25 engineers while maintaining exceptional talent density

Architected and scaled Appwrite Cloud infrastructure to efficiently handle over 100M daily requests with 99.99% availability

Spearheaded Appwrite's AI initiatives including the development of an intelligent AI assistant powered by RAG and Appwrite's documentation alongside the Model Context Protocol server

Established robust engineering principles and standards to ensure strong technical foundations and exceptional code quality across the organization

Designed and implemented comprehensive disaster recovery plans for Appwrite Cloud with multi-region failover capabilities

Led the successful migration from Docker Swarm to Kubernetes with zero downtime, improving scalability and deployment efficiency

Implemented advanced monitoring and telemetry systems using Grafana and Prometheus, enabling proactive issue detection and resolution

Orchestrated cross-functional collaboration between engineering, product, and stakeholder teams through efficient sprint planning and cycle management

Provide strategic technical consultations to C-suite executives and venture capital firms on emerging technologies, helping shape multi-million dollar investment decisions and technology adoption strategies

Conduct comprehensive technical due diligence and architecture evaluations for high-stakes acquisitions and investments, identifying critical risks and opportunities that influence strategic business decisions

Deliver expert assessments on bleeding-edge technology trends including AI/ML, cloud infrastructure, and enterprise software, enabling clients to navigate rapidly evolving technological landscapes with confidence

Architected and developed cross-platform mobile applications (iOS and Android) for Intuit Expert Platform enabling financial professionals to schedule, manage, and track client appointments, resulting in 60% improvement in productivity and response times

Engineered mission-critical backend services with SpringBoot and asynchronous processing using Kafka to support high-throughput data operations for expert-client interactions

Implemented sophisticated fault-tolerance mechanisms including circuit breakers, retry strategies, and graceful degradation to ensure application resilience during peak tax seasons

Deployed and maintained microservices in a container-orchestrated environment using Kubernetes with comprehensive monitoring, logging, and alerting to ensure 99.9% uptime

Collaborated with UX designers and product managers to continuously refine the expert experience through iterative development and user feedback loops

Architected an advanced deep learning-based image denoising system for ultra-low light photography that achieved 15% quality improvement over Samsung's production pipeline, enhancing night photography capabilities

Engineered novel neural network architecture modifications that surpassed state-of-the-art models by 7 dB PSNR, significantly improving perceptual quality through sophisticated fine-tuning techniques and custom loss functions

Conducted extensive hyperparameter optimization experiments to identify optimal training configurations, resulting in faster convergence and superior model generalization across diverse lighting conditions

Developed a sophisticated Python-based CLI automation tool that reduced environment setup time from minutes to under 5 seconds, dramatically improving team productivity and standardizing development workflows

Collaborated with senior researchers to implement and evaluate multiple state-of-the-art denoising architectures including U-Nets, transformers, and attention-based mechanisms for computational photography applications

Architected and implemented advanced computer vision APIs for Driver Drowsiness Detection that reduced accident risk by 78% in controlled trials, leveraging CNN-based facial landmark detection

Engineered sophisticated model optimization techniques that enabled real-time inference on resource-constrained edge devices like Raspberry Pi 3, achieving sub-100ms latency while maintaining accuracy

Developed a comprehensive Android native application with real-time alerting system and companion REST API backend for telemetry logging, statistical analysis, and driver performance tracking

Pioneered a real-time Automatic Number Plate Recognition (ANPR) system for Indian license plates achieving 97% accuracy despite challenging variations in plate formats, fonts, and environmental conditions

Implemented custom OCR algorithms and image pre-processing pipelines that could handle variable lighting conditions, skewed angles, and diverse regional plate formats in CCTV footage