Senior Embedded Firmware Engineer

What We're Building

Construction, a $2T\+ industry, loses hundreds of billions of dollars a year in productivity because no one can see what’s really happening on site. It’s the largest blind spot in physical intelligence.

Ironsite is building the intelligence layer for the physical world. We put wearable cameras on craft workers and use vision-language models to understand how physical work gets done. Our insights show builders where time is lost and what to change, so projects finish faster and crews spend more of their day on skilled work. We’re building the foundation models for physical work, in an industry that has never had ground truth.

The Challenge

Job sites are among the hardest places in the world to build technology. Many start as bare dirt with no power or network, and change every day as the project moves forward. Our devices and network infrastructure are field-tested in extreme weather and harsh conditions, across a fleet scaling to thousands. Every night, 1,000\+ hours of footage arrive, and by 5am it’s labeled by our own model, with our in-house experts reviewing anything it’s unsure of. Frontier models that solve research-level math still can’t count the blocks a mason laid over a 10-hour day. Scaled to 100,000 cameras, we’d record more video every day than the world uploads to YouTube.

Where We're Going

No one else has this data. Every day, our cameras record skilled trades doing real work, with 100\+ on our largest site alone. Our own model already matches expert human labelers. Now we're building world models that predict what happens on a site tomorrow, and after that, robotics. What you build this week runs on real job sites next week.

The Role

We're hiring a Senior Embedded Firmware Engineer to help own the firmware that runs Ironsite's wearable hardware.

This is a foundational role on the hardware team. The firmware you build has to reliably capture, compress, and upload full days of high-quality first-person video from active jobsites, on ruggedized wearable devices operating in the harshest environments in American industry. Battery life, image quality, thermal performance, and upload reliability are all firmware problems in this role. If the firmware breaks, the whole product breaks.

You'll work directly with our hardware, manufacturing, and cloud infrastructure teams to bring devices from initial silicon bring-up through volume manufacturing. You'll own the camera pipeline, the sensor stack, the OS, the OTA and provisioning infrastructure, and the tuning that makes footage from a hardhat-mounted camera in a dusty jobsite look and behave as good as anything shot in a controlled environment.

This is one of the most technically demanding roles in the company, and one of the highest-leverage.

What You'll Build

Firmware for our wearable hardware platform.

• Develop embedded software for our wearable devices on platforms like STM32 and Nordic, running Zephyr RTOS.
• Write and maintain drivers and libraries for camera, audio, wireless, IMU, and environmental sensors.
• Own end-to-end sensor bring-up on new hardware revisions, from initial silicon through validated production release.
• Optimize for the constraints that matter in a wearable running an 8-12 hour shift: memory, compute, thermal, and battery.

The camera pipeline.

• Own the imaging pipeline end to end, from sensor driver through ISP tuning to encoded video output.
• Tune ISP parameters (demosaicing, color correction, lens shading correction, spatial and temporal noise reduction, HDR/WDR) for the specific lighting and environmental challenges of construction jobsites, including bright outdoor sun, dark building interiors, and everything in between.
• Analyze and balance image quality tradeoffs: sharpness, dynamic range, noise, artifacts, and encoder efficiency, at the resolutions and framerates our data and AI teams depend on.
• Validate image quality across diverse lighting, motion, and environmental conditions, and build the test infrastructure that makes IQ regressions catchable before they ship.

On-device intelligence.

• Run real-time algorithms and lightweight ML inference on-device, including motion detection, event triggering, and preprocessing that reduces upstream compute and bandwidth costs.
• Partner with our AI research team to move selected inference workloads from the cloud to the edge as models mature.

Reliability, provisioning, and OTA.

• Build reliable, secure infrastructure for device provisioning, over-the-air firmware updates, and telemetry across a fleet growing from thousands to hundreds of thousands of devices.
• Design the firmware architecture and test infrastructure that makes fleet updates safe, staged, and recoverable.
• Own the debugging tools, logging infrastructure, and observability that let us diagnose issues on devices deployed in the field.

Hardware and manufacturing partnership.

• Collaborate with hardware and manufacturing engineers across the full product lifecycle, from initial architecture and prototype bring-up through DFM, DFT, and volume production.
• Design firmware alongside manufacturing test procedures. Programming stations, functional test, and factory calibration are firmware problems as much as manufacturing problems.
• Support new hardware revisions with rapid firmware iterations, and ensure firmware releases stay tightly aligned with the manufacturing cadence.

What We're Looking For

Required

• 6\+ years of hands-on experience shipping embedded firmware on production hardware, with at least one full product lifecycle delivered in wearables, cameras, or IoT devices at real volume.
• Deep expertise in C and C\+\+ for embedded systems with real resource constraints. You've written firmware that lived within tight memory, compute, and power budgets.
• Deep experience with camera and imaging pipelines, including sensor drivers, ISP tuning, and image quality analysis. This is not optional. You've owned an image pipeline in production and can speak concretely about what you tuned, what you fixed, and what tradeoffs you made.
• Strong understanding of ISP core blocks: demosaicing, color correction, lens shading correction, spatial and temporal noise reduction, and HDR/WDR performance.
• Hands-on experience with at least one industry-standard microcontroller platform (STM32, nRF, ESP32, or similar) and one embedded RTOS (Zephyr, FreeRTOS, or similar).
• Comfortable reading schematics and datasheets, and using standard hardware lab tools (oscilloscope, logic analyzer, JTAG, protocol analyzers) to debug hardware-software integration issues.
• Systems thinking across the full embedded stack. From silicon to the cloud, you understand how firmware decisions propagate up and down.
• A background in Electrical Engineering, Computer Engineering, or a related field, or the equivalent experience.

Strongly preferred

• Experience shipping wearable products at consumer volumes (Fitbit, Oura, Whoop, GoPro, Bose, Apple, Meta, Snap, or similar).
• Experience with wireless connectivity stacks (BLE, Wi-Fi, LTE, or similar) and their power and reliability implications.
• Experience with power management design and low-power firmware techniques.
• Experience shipping OTA update systems for large device fleets.

Nice to have

• Experience with on-device ML inference, including quantization, model deployment on edge accelerators, and inference optimization for constrained platforms.
• Experience with high-speed video encoding pipelines (H.264, H.265, AV1) and the codec-side tradeoffs.
• Experience with audio processing pipelines (even though we don't capture audio on our devices, adjacent experience often transfers).
• Experience contributing to Zephyr OS upstream or maintaining fork-level customizations.

What Success Looks Like

• First 90 days.

You know our hardware platform cold, including the sensor stack, the ISP, the RTOS configuration, and the manufacturing test infrastructure. You've shipped your first meaningful firmware improvement to production devices.

• First 6 months.

You've owned a major firmware initiative end to end, whether it's a next-generation camera bring-up, a fleet-wide OTA architecture improvement, or a battery life optimization that materially extends shift capacity. Your work is running on every deployed device.

• First 12 months.

You're the internal authority on firmware at Ironsite. Our devices are more reliable, more efficient, and easier to update because of you. When we scale from thousands of devices to hundreds of thousands, we do it because the firmware you built holds up.

Location, Compensation, \& Perks

• San Francisco Bay Area (on-site)
• Base salary: $200k-$300k per year, commensurate with experience
• Significant early-stage equity
• Full benefits including health, dental, vision, and 401(k) with 6% match
• Daily catered lunch
• Office in San Francisco, next to Oracle Park and the Caltrain

Final compensation is determined by experience, location, and level.

Why This Role

Every category-defining hardware company has a firmware engineer who was the reason the product actually worked in the real world. Not because they wrote the fanciest code, but because they cared obsessively about the last 5% of image quality, the last 30 minutes of battery life, and the one edge case that would have caused a fleet-wide failure at scale.

Ironsite is at that moment. We have the product, the customers, the funding, the manufacturing partnerships, and a hardware roadmap that scales into hundreds of thousands of units. What we need is the firmware engineer who is going to make sure the devices actually deliver on what our data, AI, and customer promises require.

If you've spent your career shipping camera firmware on wearables, and you're ready to do it again on a product that matters more than most, we want to talk.

Compensation

The base pay range for this role is $200,000 – $300,000 per year.