Michael Spieler
Perception Software Engineer at Lyte
Bucharest, Romania
Summary
👤
Senior🎓
Top SchoolMichael Spieler is a perception software engineer with 11 years of experience building embedded and robotics systems, currently focused on perception at Lyte in Bucharest. He brings deep expertise in C++, real-time systems, and optimal control from roles at ANYbotics, Leica Geosystems, and ETH Zurich’s Robotic Systems Lab, where he developed MPC solvers and motor control stacks. Michael contributed to the open-source ocs2 project, improving ContinuousTimeRiccatiEquations and Gauss-Newton DDP implementations for switched systems, reflecting a strong background in nonlinear optimization. He has a practical track record of hardening real-time code—finding race conditions and migrating stacks from ROS1 to ROS2—combined with hands-on embedded work on QNX and CI/TDD workflows. Trained at EPFL in microengineering and robotics, he blends academic rigor with startup grit from co-founding WISE Robotics. Colleagues rely on him for making complex perception and control algorithms robust enough for real-world deployment.
11 years of coding experience
8 years of employment as a software developer
Bachelor of Science - BS, Microengineering, Bachelor of Science - BS, Microengineering at Ecole polytechnique fédérale de LausanneMaster of Science - MS, Microengineering - Robotics, Master of Science - MS, Microengineering - Robotics at EPFL (École polytechnique fédérale de Lausanne)
German, English, French
Github Skills (9)
controls10
control-theory10
c-language10
cprogramming-language10
controlling10
linear-algebra10
control-flow10
controlled10
numerical-optimization9
Programming languages (5)
Github contributions (5)
leggedrobotics/ocs2
Oct 2019 - Jul 2021
Optimal Control for Switched Systems
Role in this project:
Back-end Developer
Back-end DeveloperContributions:562 commits, 1 comment in 1 year 9 months
Contributions summary:Michael contributed to the optimal control of switched systems by implementing changes related to the ContinuousTimeRiccatiEquations, GaussNewtonDDP, and SLQ classes. Their work focused on improving the performance and functionality by modifying functions to transcribe and convert matrices and vectors as well as calculations to determine the jump map and flow map, and testing for their correctness. Additionally, they were involved in refining code through review changes, focusing on correcting issues with the data structure used for state input.
roboticsmpcswitched-systemsoptimal-controlyarp
msplr/dotfiles
Mar 2016 - Aug 2024
Contributions:13 pushes, 1 branch in 8 years 6 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.