Working Student - Fleet Observability & Data Visualization

At Gravis, we operate at the intersection of hardware, software, and real-world deployment. Our Rooftop Autonomous Control Kit (RACK) integrates sensing, compute, communication, and networking into a manufacturer-agnostic solution deployable across a wide range of construction machines., * Build and maintain pipelines to fetch metrics from machines deployed in the field

• Organize and structure operational data so it can be visualized
• Design and implement live dashboards on our office display to distribute the information company wide
• Set up automated analysis of trends and long-term averages across different metrics
• Develop anomaly-detection workflows to flag decalibration, sensor drift, or unexpected behavior before it becomes an incident
• Leverage AI agents to accelerate data exploration, summarization, and routine analysis tasks

Do you have experience in Sensors?, Do you have a Master's degree?, We are looking for a motivated working student to join our team for an extended engagement (several months or longer), contributing 1 to 2 days per week. Our office is located in Zurich city, close to Bahnhof Oerlikon, and is easily reachable by public transport. You will own the pipeline that turns raw telemetry from our deployed machines into actionable insight - building dashboards that the whole company can read at a glance, and automating the analysis that catches problems before they reach the field engineers. This is a great opportunity to shape how a fast-growing robotics company sees its own fleet., * EU/EFTA students

• Currently enrolled at a Swiss university (Bachelor's or Master's level), ideally in computer science, data science, electrical engineering, robotics, or a related field
• Proficiency in Linux systems and comfort working from the command line
• Experience with observability and visualization tools such as Grafana, Prometheus, Foxglove, Kibana, Datadog, or comparable stacks
• Working knowledge of modern web technologies (HTML/CSS, JavaScript/TypeScript, and a frontend framework such as React, Vue, or Svelte)
• Working knowledge of git
• Reliable, detail-oriented, and comfortable owning a project end-to-end
• Good communication skills in English, * Experience with time-series databases (Prometheus, InfluxDB, TimescaleDB, VictoriaMetrics, or similar)
• Familiarity with statistical methods for anomaly detection or trend analysis Experience with data pipelines, message brokers, or stream processing (MQTT, Kafka, Zenoh, ROS bags, or similar)
• Familiarity with robotics concepts (sensors, coordinate frames, calibration, etc.)
• Practical experience using AI agents or LLM-based tools in a development or analysis workflow
• Prior experience building dashboards or monitoring systems (ideally in production)
• Long-term engagement preferred, with the possibility to grow your responsibilities over time

This is an opportunity to join a dynamic and versatile team, and to be part of a young startup that will revolutionize heavy construction. You will get the chance to apply your skills to critical tasks, while learning from a team of world-class robotic engineers. Gravis Robotics offers a fair market salary and an excellent working location in the city of Zurich.

• Hands-on experience with the full data lifecycle of a production robotics fleet - from raw telemetry to actionable insight
• Direct visibility into how a deep-tech startup uses data to scale up its operations
• The opportunity to take ownership of a system that the whole company relies on every day

Gravis Robotics is a startup turning heavy construction machines into intelligent and autonomous robots. Our unique combination of learning-based automation and augmented remote control enables a single operator to safely manage a fleet of earthmoving machines in a gamified environment. With over a decade of academic experience at the cutting edge of large-scale robotics, our team is rapidly translating this expertise into real-world deployments with industry leaders in a trillion-dollar market.