Robotics Software Engineer

Atomic Semi • Austin, TX

1d • $125,000 - $195,000 • Onsite

About The Position

Atomic Semi is building a small, fast semiconductor fab. It’s already possible to build this with today’s technology and a few simplifications. We’ll build the tools ourselves so we can quickly iterate and improve. We’re building a small team of exceptional, hands-on engineers to make this happen. Mechanical, electrical, hardware, computer, and process. We’ll own the stack from atoms to architecture. Our team is optimistic about the future and we want to continue pushing the limits of technology. Smaller is better. Faster is better. Building it ourselves is better. We believe our team and lab can build anything. We’ve set up 3D printers, a wide array of microscopes, e-beam writers, general fabrication equipment - and whatever is missing, we’ll just invent along the way. Atomic was founded by Sam Zeloof and Jim Keller. Sam is best known for making chips in his garage, and Jim has been a leader in the semiconductor industry for the past 40 years. The Tool Software team builds the software that makes our semiconductor fabrication tools run, spanning everything from microcontroller firmware to Linux applications, assembly, calibration, and over-the-air deployment. We don't silo our engineers into narrow sub-domains. Because sensing, control, and physical hardware are deeply intertwined, you will have broad control over the design of the system from top to bottom. As a Robotics Software Engineer, you will build the algorithmically rich software that brings our fabrication tools to life. You will tackle deep technical challenges that demand sub-micron positioning, non-standard sensor modalities, and tight real-time constraints—fusing multi-sensor data into a single state estimate, optimizing trajectories under real mechanical constraints, and building computer vision pipelines for alignment and inspection. We do not draw sharp lines between "software problems" and "hardware problems" because the challenges we face rarely respect those boundaries. You will spend most of your time writing code, but you will also chase bugs across domains when needed. Whether a drifting sensor turns out to be thermal coupling from a nearby motor, or a non-converging controller is actually a mechanical backlash issue, you will follow the problem wherever it leads. We are open to a range of experience levels, from exceptional early-career engineers to staff-level builders. What matters most is that you are nimble, you understand how mechanical systems actually work, and you can seamlessly bridge the gap between pure code and real-world hardware. You will know you are successful when you have built fast, highly reliable software that solves these deep physical challenges and allows our tools to operate reliably. A portfolio or Github is generally required to apply: show us the things you’ve actually built! For us, a strong portfolio demonstrates your engineering excellence, curiosity, and a love for taking ideas from concept to reality.

Requirements

• Bachelor’s degree or equivalent experience in a related field
• Strong programming skills in a systems language (C++, C, or Rust) and scripting experience in Python
• Proven experience shipping robotics or control software that ran on real, physical hardware — with attention to testing, logging, and failure modes
• Depth in at least two of: computer vision, state estimation/sensor fusion, motion control, trajectory optimization, calibration, systems modeling
• Ability to debug across domains, from high-level code down into firmware and electrical hardware

Nice To Haves

• Familiarity with precision motion — piezo stages, air bearings, flexures, or sub-micron positioning
• Experience with modern optimization frameworks, algorithms, cost-function tuning
• Experience with ML or RL frameworks applied to perception or control
• Experience with GPU-accelerated CV or real-time vision pipelines
• Basic EE lab chops: reading a schematic, using a scope or logic analyzer, chasing a noisy signal
• Experience with physics-based simulation (MuJoCo, Drake, Gazebo, Bullet, or custom)

Responsibilities

• Design and implement computer vision pipelines for alignment, metrology/inspection, object detection etc across our tools
• Build motion control and trajectory optimization for multi-axis systems, accounting for mechanical realities like backlash, compliance, and actuator nonlinearity
• Develop state estimation and sensor fusion routines that combine data from cameras, encoders, and custom sensors into reliable real-time estimates
• Write calibration procedures end-to-end — define what "calibrated" means for a new tool, implement the routines, and validate against ground truth
• Characterize and model physical systems from scratch, and build simulations that are faithful enough to catch bugs before they hit the lab
• Debug across the full stack when a problem demands it — from high-level planner, through firmware, to signals on the board
• Work directly with mechanical, electrical, and process engineers to integrate new robotic systems into production tools

Benefits

• Medical, Dental, and Vision insurance
• Generous Paid Time Off inclusive of Holidays and Sick Time
• Visa Sponsorship
• Life and Disability Insurance
• Paid Parental Leave
• 401(k) retirement plan
• Weekly Learning & Development opportunities
• Commuter Benefits including Parking and Late Night Uber rides from the office
• Lunches daily, Dinners 3x per week, Stocked Office Kitchen with Snacks and Spindrifts

What This Job Offers

• Job Type: Full-time
• Career Level: Entry Level