Feb 2024 – Present · 2 yrs 1 mo

• 1. Led the re-architecture of Nanonets into microservices, designing fairness-based processing, rate-limiting, caching, and orchestration mechanisms to reliably scale GPU workloads and ensure seamless service integration
• 2. Drove platform stability from 97.5% to 99.99% uptime by instituting incident reviews, on-call ownership, release management and advanced detection systems thereby improving SLA adherence
• 3. Directed DevOps and infrastructure initiatives, enhancing Kubernetes and CI/CD pipelines through optimized Dockerfiles, automated deployments, and robust auto-scaling, monitoring, and alerting frameworks
• 4. Designed a custom queuing architecture for GPU-intensive ML APIs to guarantee fairness and platform reliability in asynchronous processing
• 5. As a leader, mentored junior/senior engineers, established release and QA processes, and conducted 100+ technical interviews to scale engineering teams.

Jan 2022 – Jan 2024 · 2 yrs

• Developed Nanonets On-Premise enterprise product from scratch with offline micro-service architecture. Played a key role in developing and scaling core platform features.

Oct 2020 – Dec 2021 · 1 yr 2 mos

• Developed Nanonets in-house OCR engine from scratch, powering all on-prem deployments. Benchmarked the model competitively against Google Vision and AWS Textract on generic datasets.