Member of Technical Staff - Embedded Systems Engineer

Domera Labs builds Ami, the first AI companion for seniors. Ami brings the power of a smartphone into a more human form: voice-first and deeply personalized. We at Domera Labs are rethinking human-machine interaction from first principles by introducing a better interface to interact with technology: voice. By combining the most intuitive hardware design with frontier AI, Ami makes AI accessible for the fastest growing demographic worldwide. The goal is to make the everyday life of older adults more seamless while addressing that generation's biggest challenge: loneliness and isolation.

The underlying technology of Ami combines a model-agnostic streaming pipeline consisting of proprietary models and off-the-shelf frontier models to create a hyper-personalized and proactive user experience. The backbone of Ami is built by our foundational lab that builds a social intelligence model optimized for social interaction, human behaviour, and interpersonal skills.

Core Responsibilities:

• You own the device: everything between the silicon and the cloud. Our users are 75 years old and there is no screen, no settings menu, and no IT support. If the device does not just work, it fails. Your job is to make that true.
• Device identity and fleet management: provisioning, unique device identity, secure key storage, and the systems that let us know what every unit in the field is doing.
• The maintenance channel: a reliable service and diagnostics path to the device (BLE, USB-C, or both) for field debugging, factory test, and recovery.
• Out-of-the-box experience: first-boot network onboarding that a senior can complete alone, with no screen and no smartphone as a crutch.
• Flash and update management: partition layout, A/B OTA with rollback, staged fleet rollouts, and bricking as a category of bug that never happens.
• Base platform behaviours: boot and recovery, factory and soft resets, button and LED mappings, audio feedback, power states, and battery behaviour.
• The audio front end: working with our ODM and silicon partners on mic array, echo cancellation, wake word, and the latency budget from mic to first spoken token.
• Bringing the device through EVT, DVT, and PVT with our ODM as the technical counterpart on our side.

What we are looking for:

• A builder's mindset: ship > fail > learn > improve.
• Stay focused when multiple priorities compete and the ranking is unclear.
• Work seamlessly across research and engineering.
• Can navigate ambiguity and make progress in fast-moving research environments.

This is a founding-level position. You will work directly with the founders, shape the architecture from the ground up, and see your work in users' hands within months, not years.

Technical Requirements:

• You have shipped a connected consumer device to production, not just to a demo table. You know what goes wrong between DVT and mass production.
• Strong C/C++ and Linux-based embedded platforms (Yocto, Buildroot, or similar), plus comfort in Python for tooling and test.
• Wireless connectivity in the real world: Wi-Fi provisioning, BLE, and ideally cellular. You have debugged a device that would not join a network in someone's home.
• OTA and flash: bootloaders, secure boot, A/B partitioning, rollback, and staged rollouts across a fleet.
• You debug with a scope and a logic analyser when the logs run out.
• A builder's mindset: you'd rather ship and measure than debate.
• Bonus: real-time audio pipelines (I2S, DSP, AEC, wake word), on-device ML inference, factory test and manufacturing line bring-up, regulatory certification (CE, FCC), power optimisation for battery life, experience in NLU / NLP.

We consider candidates across experience levels and lean toward senior engineers who have worked directly with frontier AI systems. Python, deep learning frameworks, or distributed computing are non-negotiable for this role.

What you get:

• Founding-team equity and a real say in technical direction.
• A product with signed launch partners, not a demo in search of a market.
• A small team, short decision paths.
• Cash compensation range of CHF 80 - 100k with significant equity incentives.
• Opportunity to shape the future of hardware and social intelligent AI models.