Electrical Engineer, Sensing

Weave Robotics • San Francisco, CA

7d

About The Position

Weave was founded to build the robots we’d want to have in our own home. We believe the next generation of robotics will transform everyday life by enabling people to do more and to reclaim time to spend on what’s important. We also believe robots are in a sense like any other product: to matter, they have to ship. Our robots are already operating in real homes and businesses, giving us the opportunity to rapidly improve from real-world experience. With a growing team, strong customer demand, and capital for expansion, we’re entering an exciting stage of growth—and we’re looking for people with exceptional talent and standards to help bring home robotics to millions of households. The Role Weave builds and deploys some of the first bimanual robots operating in customer homes. Our robots carry a suite of sensors beyond their cameras that keep them capable and safe around people. As a Sensing Engineer, you'll own these sensors end-to-end: the emitter and detector electronics, the analog front ends, and the processing and calibration that make them dependable for the firmware, software, and autonomy layers above. The role spans analog design and signal processing, and it isn't scoped to specific modalities. Modalities will change between robot generations (optical, magnetic, capacitive), and sensors end up embedded in soft materials, woven into fabric, or mounted next to motors. The job is taking any of them from evaluation through production.

Requirements

• End-to-end sensor ownership: 3+ years spanning sensing hardware and signal processing, with at least one sensor you've carried from schematic through shipped, calibrated signal.
• Mixed-signal competence: analog front-end design, signal conditioning, grounding and shielding judgment, and the layout instincts that keep microvolts alive next to amps, across dozens of channels, not one.
• Signal processing and estimation fundamentals: filtering, spectral analysis, and state estimation applied with judgment, including inverse problems where the quantity you want is a nonlinear function of what you can measure, plus the physics literacy to understand why a sensor misbehaves, not just that it does.
• Experimental discipline: you design the measurement before the algorithm, and your claims come with error bars.
• Modality range: you've worked with more than one sensing modality, or can show you learn the physics of a new one fast.
• Communication protocols: EtherCAT, CAN/CAN FD, SPI, and I²C.

Nice To Haves

• Depth in a specific modality: optical emitter/detector chains (TIAs, ambient rejection, modulated ranging), magnetic sensing (magnetometer arrays, field modeling), capacitive or resistive sensing in textiles, force, inertial, or acoustic.
• Flex-PCB, electrode, or sensor-in-mechanism design with mechanical teams: sensing that lives inside soft or moving structures.
• Production calibration: including fixture design and station software.
• Sensors feeding closed-loop control: with contractual latency and noise budgets.
• Laser eye-safety classification and compliance.
• Fluency in Rust/Python and C/C++: analysis and prototyping in Python, production implementations that run in real time on embedded compute.

Responsibilities

• Own sensors end-to-end: architecture, electronics, sampling, signal processing, calibration, and production test for the robot's time-series sensing, with sensor selection driven by what the robot needs to feel.
• Design sensing electronics: analog front ends, emitter drive and detector receive chains, mixed-signal circuits, and multi-channel arrays (multiplexing, sampling synchronization, and timestamping tight enough for the state estimators that consume them), delivering clean signals with switching power stages, motors, and moving structures all nearby.
• Write the algorithms: filtering, bias and temperature compensation, drift handling, outlier rejection, and the detection and estimation layers that turn raw samples into quantities other teams can trust, including inverse models that recover contact and geometry from what's actually measurable.
• Characterize everything: design fixtures and experiments that map each sensor's real behavior (noise, drift, crosstalk, hysteresis, latency, and failure modes) and turn those maps into models the algorithms use.
• Build calibration for production: per-unit calibration routines that run on our manufacturing line and in the field, so every robot's sensors agree with reality, not just the prototype's.
• Detect sensor failure: health monitoring that distinguishes a degraded sensor from a changed world, with honest confidence estimates attached to every signal.
• Evaluate what's next: prototype candidate modalities quickly, and give hardware and product teams data-backed answers about what earns a place on the robot.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Education Level: No Education Listed