Embedded Engineer

Accelerating embodied AGI. The next ten years of AI will not be won in software. They will be won in the physical world: in factories, hospitals, kitchens, fields, and homes. The companies that own that data will own the century. MicroAGI is building it. We are a data research and deployment lab. Atlas captures the manual work inside Europe’s industrial champions, trains foundation models on it, and deploys the right robot onto the floor, with no teleoperation and no interruption to the line. Shift records how physical work actually happens and turns it into training-ready data. We were founded by former Formula 1 engineers from Red Bull Racing and Mercedes, together with a researcher from the Alan Turing Institute. Our $55M seed round, the largest in German history, was led by Hummingbird with Northzone, LocalGlobe, Village Global, and redalpine. We are based in Munich, with our research centre in Zürich and teams in London, New York, and Istanbul. We are now building the engineering and research teams that will take Atlas into new industries. Our models learn from data captured in the physical world, by hardware we design and build ourselves. As Embedded Engineer, you will own the firmware that runs on that hardware, from the first register write to the moment data leaves the device, so that what reaches our ML team is clean, synchronized, and trustworthy. What You Will Do Write and own the firmware for our data-capture hardware: sensors, motion-capture rigs, custom IoT setups, and prototype robots. Build real-time data acquisition: drivers, DMA, buffering, and streaming from many sensors at once without dropping a sample. Solve multi-sensor synchronization, calibration, and time-stamping so every data point lines up across devices and modalities. Bring up new boards with our Electrical Engineer and turn fresh hardware into working devices fast. Build the path from device to dataset: communication protocols, on-device logging, and handoff into our data pipeline. Work shoulder to shoulder with ML engineers. Your firmware decides what their training data looks like, and a timing bug can quietly ruin months of data. Make firmware reliable in the field: testing, OTA updates, error handling, and diagnostics that tell us what went wrong. Requirements Up to 5 years writing production firmware for embedded or IoT systems. Strong C/C++, experience with an RTOS (FreeRTOS, Zephyr, or similar), and deep familiarity with microcontrollers (STM32, ESP32, NXP, or similar). Hands-on experience with peripheral interfaces and protocols: SPI, I2C, UART, USB, BLE, CAN, or similar. Experience integrating sensors in firmware: IMUs, cameras, depth sensors, force sensors, or similar. A strong understanding of how firmware decisions affect data quality, especially for EMG/EEG sensing: ADC configuration, sample rate, jitter, filtering, and buffering. You know that a great model on a noisy signal still loses to a decent model on a clean one, and you write code accordingly. Comfortable debugging at the hardware boundary with a logic analyzer, oscilloscope, and debugger. Genuinely curious about ML: you read papers, you have trained at least one model end to end, and you understand that dataset quality beats model architecture choice. High agency: you don't wait to be told what to do. You are comfortable in 0-to-1 ambiguity, where the spec is "figure out what the spec should be". Discreet: you handle confidential hardware roadmaps and customer data with care. Fluent in English. German is a plus. Nice to Have A background in robotics, motion capture, or wearable devices. Experience with real-time or hard-deadline systems: audio, video sync, or robotic control loops. Experience with embedded Linux, time-sync protocols (PTP or similar), or low-power design. Familiarity with ML training pipelines. Even if you don't run them, you can read a training script. Open-source firmware contributions. Experience taking firmware from prototype to small-batch production: test fixtures, provisioning, and certification support. Who Will Thrive at MicroAGI Everyone we hire is assessed on the same four traits: Mission and vision: you care about what we are building and why it matters, and your past choices show it. Intensity: you work at a pace that matches the size of the opportunity, and you sustain it. Entrepreneurial: you start things nobody asked for, and you finish them. Raw IQ: you learn fast, reason from first principles, and get sharper when the problem changes. This is for you if: You would rather own a problem end to end than wait for a spec. You use AI every day, and you can tell a correct answer from a plausible one. You want to work in person, in the office, with people who are very good at what they do. This is not for you if: You are looking for a predictable 9-to-5. You would forward output you cannot explain. "That’s not my job" feels like a reasonable answer. How We Hire We read every application and reply within five working days. Talent Screen (30 min): a first conversation, both ways. Technical or Functional Interview (60 min): how you think when the problem changes. Your Work (60 min): bring something you are proud of and walk us through it. Leadership (30 min): a conversation with our CEO. Coworking Day (engineering and research roles): a day building with the team on a real problem. Then a decision, usually within a few days. Pay Transparency Compensation includes base salary and equity. The salary range for this role is shown on this posting. What We Offer Agency: real ownership from day one, and decisions that actually matter. Freedom: to choose how you solve the problem, without layers of approval. A seat on a rocketship: one of Europe’s fastest-growing deep-tech companies, at the earliest stage. Free food and drinks in the office. Equal Opportunity MicroAGI welcomes applications from everyone. We assess candidates on their work and potential, regardless of ethnic origin, gender, religion or belief, disability, age, or sexual identity. Find more English Speaking Jobs in Switzerland on Arbeitnow