Who we are
At Jaipur Robotics, we're building AI-powered solutions that transform the waste industry — turning complex data into automation, efficiency, and environmental impact. As a fast-growing, VC-backed clean-tech startup based in Switzerland, we work with leading waste operators across Europe and are scaling fast in the field.
What we offer
We're hiring a Senior MLOps Engineer to own the full ML lifecycle behind our AI inference products: data curation, training, deployment, inference and continuous improvement.
This is not a DevOps-only role. You will own multiple computer vision inference projects on GCP end to end: working out what data is needed, training and deploying models, and monitoring performance and degradation to feed new data back into training. We want a bias to action: you ship, measure and iterate without waiting for a handoff.
- Work directly with founders and R&D engineers on core systems
- Contribute to scaling real-world AI systems used in industrial environments
- Competitive salary and ESOP
Key Responsibilities
Data Curation, Training & Project Ownership
- Own each AI inference project end to end: work out what data is needed, then drive it through training, deployment and live inference
- Define data requirements with R&D engineers and founders; source, curate, label and quality-check datasets (including edge cases), then train, fine-tune and evaluate models
MLOps & CI/CD
- Build and maintain CI/CD pipelines (GitHub Actions) that automate model training, validation and deployment
- Implement safe releases: staged rollouts, rollbacks and automated evaluation gates before a model ships
- Version datasets, models and pipelines with lineage so every result is reproducible
Cloud, Data & Embeddings
- Deploy and manage services on GCP (Cloud Run, GKE, GCS, Pub/Sub) using Docker and Kubernetes
- Build scalable batch and streaming pipelines (Apache Beam / Dataflow) for large image and sensor datasets
- Store and serve embeddings with vector databases and similarity search
Inference, Monitoring & Reliability
- Monitor model performance, accuracy degradation, drift, services and cost, with alerting and observability
- Deploy and operate inference services in production, ensuring correct output and the latency and throughput each project needs
- Close the loop: use monitoring to find failure cases and data gaps, and feed new data into retraining
- Build MCP (Model Context Protocol) servers and tooling that let AI agents query data, embeddings and evaluation results
Requirements
- Strong experience with GCP (Cloud Run, GKE, GCS, Pub/Sub, IAM)
- Experience with CI/CD (GitHub Actions or similar), Docker and Kubernetes in production
- Experience building batch and streaming data pipelines (Apache Beam / Dataflow)
- Experience with dataset and model versioning and automated evaluation pipelines, plus embeddings and vector databases
- Hands-on, end-to-end ML lifecycle experience (data curation, training, deployment, inference, monitoring), including production failure handling and debugging
- Experience training or fine-tuning ML models, ideally computer vision (e.g. PyTorch)
- Ability to run multiple AI inference projects in parallel
- Strong programming skills in Python
Nice to Have
- Experience working in an early-stage startup / scale-up (<50 engineers)
- Experience with camera and LiDAR systems, or edge deployments in restricted IT/OT industrial environments
- Terraform (infrastructure-as-code)