Headline

SDE-1 @ Ola Krutrim | Backend Engineer | Java · Spring Boot · Distributed Systems · Generative AI | NIT Patna ’25 Knight 1704

About

I build things that don't break at 3am. Currently at OLA Krutrim as a Backend SDE-1, I work on production cloud-native infrastructure — NUMA-aware VM scheduling on OpenStack, VPC services, Load Balancers, and Auto Scaling Groups. In 6 months, I've: → Reduced memory over-provisioning by 20% → Cut API response latency by 25% → Helped bring incident escalations down by 40% → Maintained 80%+ test coverage on critical code paths These aren't resume numbers. They're what production systems teach you when things go wrong at scale. My core stack: Java · Spring Boot · Go · Microservices · Docker · Kafka · Redis · PostgreSQL · AWS · OpenStack. I'm drawn to distributed systems problems, concurrency challenges, and backend architecture decisions that matter under real load. Before Krutrim, I built a Unified Access Management system at Inventeron Technologies — optimizing MySQL queries, designing RESTful microservices, and implementing JWT + RBAC security that eliminated 100% of unauthorized access incidents post-deployment. On the side: I published research on image compression at NIT Patna under Prof. Amit Kumar Singh (Top 2% scientist globally), and led development of QuickTongue — the world's first tongue-controlled game, combining computer vision, AI, and HCI for assistive technology. Competitive coding: 1100+ LeetCode problems, rating 1704 (Knight). I solve problems because I genuinely enjoy the craft — and it shows in how I debug production issues. NIT Patna CS '25 · Actively seeking SDE-1 backend roles at product companies across India. Let's connect → deepak.singh.tech48@gmail.com

Location

Bengaluru

AI stack

anthropic apis

antigravity

claude code

claude

codex

cursor

Work experience

SDE 1 at Krutrim, Oct 2025 - Present · 1 y, Bengaluru, Karnataka, India, Building production cloud-native infrastructure at OLA Krutrim — one of India's fastest-growing AI cloud platforms. • Architected vmSmartScheduler — a NUMA-aware VM scheduling service on OpenStack Placement APIs across 20+ compute hosts, reducing memory over-provisioning by 20% • Designed and owned backend services for VPCs, Load Balancers, Virtual Machines, and Auto Scaling Groups; enforced resilience patterns maintaining 99.9% production availability • Built thread-safe backend automation using Java multithreading and event-driven workflows, reducing manual operational effort by 35% • Designed internal REST APIs and SDKs for inter-service communication between compute, networking, and orchestration layers → 25% lower API response latency • Resolved critical race conditions in OpenStack production with SRE teams → 40% reduction in incident escalation frequency • Maintained 80%+ test coverage on critical code paths using JUnit and Mockito

Java Software Developer at Inventeron Technologies, Jan 2025 - Jun 2025 · 6 mos, Bengaluru, Karnataka, India, Built and scaled backend services for a Unified Access Management (UAM) system centralizing user access across multiple client applications. • Built production backend services in Java and Spring Boot supporting 99.9% uptime across multiple client environments • Optimized MySQL schema and query execution plans → 30% faster average API response time, significantly lower p99 latency under peak load • Implemented Spring Security with JWT-based authentication and RBAC authorization → eliminated 100% of unauthorized access incidents post-deployment • Architected 3 independent microservices from scratch with CI/CD pipelines → 40% faster feature rollout for the engineering team • Collaborated in Agile sprints — active in code reviews, sprint planning, and technical design discussions

Research Intern — Image Compression & Assistive AI at National Institute of Technology , Patna, Aug 2024 - Dec 2024 · 5 mos, Patna, Bihar, India, Research project under Prof. Amit Kumar Singh (Top 2% scientist globally, NIT Patna) — focusing on real-time image compression and AI-powered assistive technology. • Designed and implemented efficient image compression algorithms to improve real-time performance and reduce storage overhead • Led development of QuickTongue — the world's first tongue-controlled game, combining computer vision, AI, and human-computer interaction for accessibility and linguistic research • Applied deep learning and OpenCV for real-time facial landmark detection and gesture recognition • Published research findings combining CV, AI, and HCI in an innovative accessibility use case

Software Developer at ITJOBXS, Jun 2023 - Aug 2023 · 3 mos, Remote, Built a full-stack Learning Management System (LMS) portal from scratch using the MERN stack. • Developed RESTful backend with Node.js and Express.js; designed MongoDB and MySQL schemas for course and user data • Implemented JWT + Bcrypt authentication with role-based access for instructors, students, and admins • Integrated real-time chat and notifications using WebSockets for live classroom interaction • Deployed on Vercel with responsive React.js frontend using component-based architecture

Education

Bachelor of Technology - BTech, Computer Science and Engineering, National Institute of Technology , Patna

Public links

github: https://github.com/deepaksinghcsebtech2025