Saurabh
Mete
I build high-throughput production systems — from real-time ML inference on edge devices to enterprise backends serving 150K+ users.
ABOUT ME
I don't just write code.
I engineer systems.
I design backend systems from the ground up — scalable, observable, and built to last. Whether it's processing real-time sensor streams at 500Hz, deploying ML models on edge devices, or architecting distributed services for enterprise logistics, I focus on software that ships and performs under pressure.
Currently at MotionMiners GmbH, I co-led an AI-powered ergonomics platform serving enterprise clients including Dachser, Škoda, Hermes, and DB Schenker. Previously at Tata Consultancy Services, I built the backend for a platform serving 150K+ users and 7K+ drivers across India.
I hold an M.Sc. in Data Science from TU Dortmund, with my thesis focused on real-time human activity recognition using on-device ML.
EXPERIENCE
Where I've built things that matter.
MotionMiners GmbH
- ●Co-led end-to-end delivery of a production edge-computing system — from ML model training to on-device inference pipeline — processing 500Hz sensor streams in live logistics environments.
- ●Trained a custom PyTorch model achieving 80%+ accuracy with only 150 min of training data; reduced model size by 50% with zero accuracy loss via quantization and pruning.
- ●Designed and built reactive data pipelines for write-heavy, high-frequency workloads on resource-constrained hardware using Kotlin Coroutines and Flows.
- ●Drove system validation through 600+ hours of user testing; iterated on performance and UX alongside product and stakeholder feedback loops.
- ●Built a high-throughput signal processing engine handling 600+ concurrent data streams in production — robust state management, fault tolerance, and sub-second latency.
- ●Delivered to enterprise clients across logistics and automotive: Dachser, Škoda, Hermes, DB Schenker.
- ●Maintained 90%+ automated test coverage; established clean architecture standards and led code review practices across the team.
Tata Consultancy Services
- ●Built backend services with Spring Boot, Hibernate, PostgreSQL, and Redis for a national transport platform serving 150K+ employees and 7K+ drivers.
- ●Designed REST APIs and data models for high-volume workflows, ensuring reliability at scale.
- ●Delivered cross-platform features: backend (Spring Boot), mobile (Java/Flutter), frontend (AngularJS).
- ●Automated regression testing with Katalon Studio; acted as Scrum Master driving sprint delivery.
EDUCATION
The foundation I built on.
Technical University Dortmund
- ●Focused on machine learning, statistical modeling, and large-scale data processing.
- ●Thesis on real-time human activity recognition using on-device ML — directly applied in production at MotionMiners.
- ●Coursework spanning deep learning, computer vision, distributed systems, and applied statistics.
University of Mumbai
- ●Strong foundations in algorithms, data structures, operating systems, networking, and database systems.
- ●Final-year project in client-side development and backend systems integration.
- ●Graduated and joined Tata Consultancy Services as Full Stack Developer.
PROJECTS
Built for production. Designed to scale.
Airgo XP — AI Ergonomics Platform
Real-time posture intelligence for industrial workers
Logistics workers face high injury risk from repetitive strain. Existing solutions required cloud connectivity and expensive multi-sensor rigs.
Self-contained edge system with a custom PyTorch inference pipeline running entirely on-device. One compact wearable classifies complex body movements in real-time at 500Hz.
Deployed to enterprise logistics environments. 80%+ ML accuracy with only 150 min of training data. 50% model size reduction with zero accuracy loss.
Companion — BLE Fleet Intelligence
Real-time asset & worker tracking at warehouse scale
High-throughput signal processing system for real-time location intelligence at warehouse scale. Ingests and correlates concurrent data streams from hundreds of devices with reliable state management and sub-second latency.
FlightRadar API
High-performance real-time flight tracking backend
Backend API for real-time flight data ingestion and WebSocket streaming. Built with Kotlin Ktor and MongoDB, deployed on AWS — designed to showcase event-driven architecture at scale.
Enterprise Transport Platform — TCS
National-scale employee transport logistics for India
Large-scale backend serving 150K+ employees and 7K+ drivers across India. Built with Spring Boot microservices, PostgreSQL, and Redis — full-stack delivery from APIs to mobile client.
SKILLS
The tools I build with.
WHY ME
What makes me different.
Not just a developer. A systems thinker who owns outcomes, not just tasks.
End-to-End Ownership
I don't hand off at the API boundary. I've shipped systems spanning ML model training, Android app architecture, backend data pipelines, and cloud deployment — fully owned from idea to production.
Performance at Scale
From 500Hz sensor streams to 150K-user backends, I've built systems that hold up under real-world pressure. Performance isn't an afterthought — it's a design constraint from day one.
Production-Proven Track Record
Airgo XP runs in live logistics environments. The TCS transport platform handles thousands of drivers daily. My work isn't demos or side projects — it's software with real users on the other end.
Cross-Domain Expertise
Backend engineer who can read a PyTorch model, design a low-latency data pipeline, and reason about hardware constraints. This rare combination means I integrate fast and reduce team friction across the stack.
Pragmatic & Fast
I ship. I've worked in fast-moving startup environments and large enterprise teams, adapting quickly, writing clean maintainable code, and keeping delivery velocity high without sacrificing quality.
Driven by Real Impact
Every system I've built has a real user — a logistics worker protected from injury, an employee getting to work safely. That context shapes every design decision I make.
CONTACT
Let's build something
impactful together.
I'm actively looking for backend engineering roles where I can own systems, drive impact, and work with teams that care about quality.