Ardi Loot
Principal Software Engineer at Microsoft
Estonia
Summary
👤
Senior🎓
Top SchoolArdi Loot is a Principal Software Engineer with nine years of experience building perception systems for autonomous vehicles, currently focused on advanced sensor fusion and detection at Microsoft after a multi-year tenure leading perception work at Starship Technologies. He holds a PhD in Physics from the University of Tartu and brings a research-first mindset to production problems, having translated experimental optics and simulation expertise into practical ToF, stereo, radar and ultrasonic perception algorithms. His work spans classical methods (multi-level clustering, Random Forests, Kalman filters) and neural pointcloud/semantic segmentation models for detecting stairs, holes, snowpiles and unmapped drivable areas, plus sensor calibration and hardware-emulating testbeds to improve reliability. Known for rewriting critical subsystems (radar) and creating machine-learned range estimators, he pairs deep analytical skills with hands-on product engineering. Active on GitHub, he maintains a portfolio that reflects breadth across algorithms, simulation and embedded-aware perception solutions.
9 years of coding experience
16 years of employment as a software developer
Hugo Treffner's High SchoolPhysics, Physics at University of TorontoDoctor of Philosophy - PhD, Physics, Doctor of Philosophy - PhD, Physics at University of Tartu
English, Estonian
Github Skills (46)
visualization10
qt10
solve10
dac10
esp826610
esp3210
rgb10
scientific-visualization10
calculation10
arduino10
wifi10
journal9
nonlinear9
python9
sum9
Programming languages (5)
Github contributions (5)
ardiloot/SMolPhot
Jul 2016 - Aug 2017
Contributions:1 release, 8 commits, 5 pushes in 1 year 1 month
ardiloot/NonlinearTMM
Feb 2017 - May 2018
Transfer-matrix method (TMM) is powerful analytical method to solve Maxwell equations in layered structures. However, standard TMM is limited by infinite plane waves (e.g no Gaussian beam excitation) and it is only limited to linear processes (i.e calculation of second-harmonic, sum-frequency, difference-frequency generation is not possible). The aim of this package is extand standard TMM to include those features. The physics of those extensions are described in the follwoing publications, first extends the standard TMM to nonlinear processes and the second extends to the beams with arbritary profiles.
Contributions:11 releases, 192 commits, 8 PRs in 1 year 3 months
aimbeamplane-wavesgaussianextensions
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.