About Human Archive

Human Archive is a research lab backed by Y Combinator focused on modeling human embodied intelligence.

Humans are the most sophisticated biological systems we have ever observed, yet we still do not fully understand ourselves. Research into human physical intelligence — including the human hand, proprioception, and vision — remains largely unsolved. Our mission is to recover human embodied intelligence as a learned model. To achieve this, we build custom hardware products, deploy them globally at scale, and publish research. Today, our data is used for robotics and world modeling, but the broader opportunity is advancing scientific research into intelligence itself.

Founded by Stanford and UC Berkeley researchers, we are lean, deeply technical, and operate at extreme speed, taking on unglamorous and conventionally impossible problems that directly unlock step-function gains in model capability.

The deployment of capable humanoids at scale will permanently redefine human labor. Undesirable physical work will disappear, and human effort will shift toward a new era of abundant creativity.

We are building the infrastructure to accelerate that transition by assembling the Human Archive mafia. You will own meaningful systems from day one and see your work directly impact model capabilities. This is a once-in-a-generation inflection point. If you want to help reshape physical labor and work on problems that matter at civilizational scale, join us.

What you'll work on
We're hiring a firmware engineer to own the embedded software stack on our wearable capture hardware. Close-to-the-metal work at the intersection of high-bandwidth sensor data, precise timing, and constrained-power embedded compute. You read schematics, debug buses, and know when the bug is in firmware versus when it's in hardware.

System architecture

• Firmware architecture on embedded Linux on ARM

• System state machines: boot, capture, idle, fault recovery

• Power management states and low-power transitions

Sensor integration and timing

• Drivers for cameras, IMUs, microphones, environmental sensors

• Sub-millisecond cross-sensor synchronization

• Hardware-level timestamping infrastructure

• Real-time sensor data alignment

• Lightweight on-device fusion for orientation and worn-state estimation (Kalman variants)

Data pipeline

• Multi-stream concurrent capture

• Hardware video encoder pipelines

• High-throughput storage writes

• Industry-standard delivery format packing

Connectivity and fleet operations

• Device-to-backend telemetry (battery, storage, session state, error events)

• Secure OTA firmware updates with signed images and rollback protection

• Remote configuration and fleet management commands

• Communication protocols for device-cloud messaging (MQTT-over-TLS, HTTPS, WebSockets)

Hardware / software bring-up

• Firmware-side board bring-up alongside electrical engineers

• Validate hardware interfaces (I²C, SPI, USB, MIPI)

• Diagnose timing, power sequencing, and communication faults

Required technical experience

• Strong C/C++ for embedded systems on ARM

• Python for testing, tooling, and automation

• Camera driver experience (V4L2 or equivalent)

• Deep familiarity with embedded interfaces: MIPI CSI-2, I²C, SPI, USB, interrupts

• Comfortable reading schematics and datasheets

Systems and concurrency

• Interrupts, race conditions, memory constraints, real-time scheduling

• RTOS concepts and multi-threaded embedded environments

Strong plus

• Qualcomm, NXP, or Ambarella BSP experience

• SerDes driver integration

• Hardware H.264 / H.265 encoder pipelines

• Precise timing and synchronization (PTP, IEEE 1588)

• High-bandwidth data acquisition

• Sensor fusion and Kalman filter implementation

• OTA infrastructure for deployed fleets

• IoT telemetry stacks (MQTT, AWS IoT, Azure IoT Hub, or similar)

• ROS, MCAP, or Foxglove

About this role Deep systems work. Long hours in the lab during board bring-up. Drivers nobody has written for the silicon you're working with. Firmware that runs unattended on real users for full shifts — and gets updated remotely without ever recalling a device.