Senior Firmware Engineer

Cairns Health is building an AI-powered care companion that seniors interact with entirely through voice. We’re looking for a highly skilled engineer to own and evolve the real-time audio and speech processing pipeline that makes these conversations feel natural, reliable, and responsive on embedded hardware.
This role is ideal for someone with a strong foundation in C++ on embedded Linux and deep hands‑on experience with audio signal processing for speech
, whether your background is firmware-heavy with ML exposure or ML-leaning with strong systems skills.

You will lead development of our on-device audio pipeline, focusing on real‑time speech interaction:

• Design and implement real-time streaming of speech audio to and from the OpenAI Realtime API
• Build and tune audio buffering, latency management, and synchronization for conversational speech
• Implement speech interruption detection (barge‑in) to support natural, turn-based dialogue
• Develop dynamic noise floor detection and related signal conditioning for in‑home environments
• Apply practical audio signal processing and ML techniques to improve speech quality and robustness
• Evaluate and potentially re‑architect our Linux audio stack (e.g., Pulse Audio → Pipe Wire)
• Optimize performance, memory usage, and reliability on constrained embedded devices
• Collaborate closely with firmware, ML, and hardware teams to ship production‑quality systems

All core development is done in C++ on Yocto‑based embedded Linux systems
.

• Strong proficiency in C++ with experience building production, real‑time systems
• Hands‑on experience with audio signal processing for speech
  , such as:
• Audio buffering and streaming
• Noise estimation / suppression
• Voice activity detection or interruption handling
• Experience developing on embedded Linux (Yocto preferred)
• Solid understanding of multi‑threaded, low‑latency systems
• Comfortable working close to the OS and audio stack

• Experience integrating with speech or conversational AI systems
• Familiarity with ML tools or models used in audio/speech processing
• Experience with Pipe Wire, Pulse Audio, ALSA
  , or similar Linux audio frameworks
• Background in embedded firmware, device bring‑up, or kernel‑adjacent development
• Experience with NXP i.MX8 or similar So Cs

Who You Might Be

• An ML or audio engineer who enjoys working close to the system in C++
• A firmware engineer who has built speech/audio pipelines and understands ML concepts
• Someone who likes solving hard, real‑world problems in noisy, unpredictable environments (homes, not labs)

Cairns Health (https://(Use the "Apply for this Job" box below).) is creating a fundamentally better healthcare experience for people with chronic health conditions and those who care for them. We make healthcare more accessible by simplifying complex care plans, connecting care teams and meeting patients where they live. Through our conversational AI, patients use their voice to interact with our digital care companion, who proactively gives medication reminders, symptom checks, behavioral nudges and even engages in friendly conversation to ease loneliness.

Cairns uses a device that includes radar to put the patient in context and passively monitors their activities, including: heart rate, breathing rate and sleep stages, all without a wearable. The result is informed and timely intervention that drives improved clinical outcomes, reduced care delivery costs and a more satisfactory healthcare experience for all.

To Apply
Please send your resume (and optionally a short note)

In office role in Sunnyvale, CA. Works 4 days in office.