Aishanou Rait, Senior Perception Engineer: “I get things done!”
In our latest employee spotlight interview, meet Aishanou Rait, one of our talented and dedicated engineers, as she talks about life at Starsky, how she became interested in a career in Robotics and challenges women face in tech. Spoiler: Aishanou’s team is growing fast and has a bunch of open positions right now!
Aishanou, tell us about your role at Starsky. What projects are you working on?
I work in the Perception team, which is responsible for taking the raw data from various sensors and converting it into a meaningful representation of the environment for Behavior planning, Local planning and Controls teams to use. Sensors include cameras, radars, IMU, and GPS. We use all that data to develop a two-dimensional view of the world, where the lanes are represented by lane lines and vehicles are represented as boxes.
I personally work on the Robotics side of things which entails post processing of the detections from the Deep Learning models, tracking, estimation, calibration and sensor fusion.
What drew you to Starsky originally and how has the company changed since?
I have been with Starsky for two-and-a-half years now. I was the seventh employee in the company. It has changed a lot since then. Initially, there were just two of us working on the Perception system and it was more of a prototype phase when we were hacking things together. Now, it’s way more structured. We have multiple projects that we are managing, and I’m a project lead.
I heard about Starsky through CMU’s network when it was still in stealth mode, and thought their business model was very promising. I always wanted to get into autonomous driving, especially in an early stage start-up to get a broader experience.
I could have worked for bigger companies, but if I had done that, I would likely be given a specific project which doesn’t capture the big picture. I was always more interested in getting things done and building a real product instead of doing something which would probably be used five years from now. I wanted to know the challenges and core problems of the technology, instead of one problem in one small part of the full product. That’s why I wanted to join an early stage start-up, and Starsky was the perfect fit.
“ I was always more interested in getting things done and building a real product instead of doing something which would probably be used five years from now.”
What do you like most about your job?
I really like learning a lot and doing things I have never done before. It’s interesting when you’re given a problem where you only know the starting point and the end goal. There is no connection in between and you are supposed to come up with a solution and connect the dots using bits of information available on that subject.
For example, you’re given a new sensor like IMU. You have to make the decision to select which make and model to use, what the best place to mount it on the vehicle is, and which information from it is most useful. You have to look at the data for that, the biases and other sources of noise. You must analyze all the issues you can face with modeling it one way versus another, learn how the vehicle dynamics work, write the code for it, and then stitch it together with the rest of the system. Hence, you are making that entire connection, figuring out each and every piece of it. You can learn so many additional things on the way up which you probably wouldn’t learn in other places and industries, because someone would tell you how to do it.
Why did you choose to go into robotics and autonomous tech in particular?
I did my master’s in Robotics Systems Development. Part of why I went into this field was that I have a tendency of getting bored with things very quickly. There is so much to learn in robotics that I don’t think I will ever get tired of it in my lifetime. So, this is a good career path, where you keep learning a lot of new, interesting things every day. It’s also fascinating to see the interaction of software with real world objects.
When I was studying at CMU, we had a project in which we functioned as a team to build a robotics system. We did a drone simulation for an underwater vehicle. Since we could not get an underwater vehicle, we simulated the entire environment. The end goal was to be able to mount the drone on a moving platform from under it. We built the entire hardware for the platform and implemented algorithms on a DJI drone. That sort of set the stage for Starsky, because working in a small team, figuring out the hard problems together and coming up with solutions under time constraints was great preparation for my current role.
“There is so much to learn in robotics that I don’t think I will ever get tired of it in my lifetime.”
Do you notice a lack of women in robotics? Why do you think that’s the case?
I think it’s more historical than anything else and the problem is compounded as a result. Women in technology were fewer to begin with and that limits the number of senior women in leadership positions which in turn limits the realization of the importance of hiring women.
It’s also a traditional thing. The gender gap is actually not that big when you get into college, but as education becomes more and more advanced, women tend to move out. After a certain age when people have families, many women tend to prioritize their families over their work. This is the time when higher leadership can help in developing policies conducive to creating a better balance.
As an engineer, I have definitely felt a lack of senior women and, as a result, lack of mentorship from female leaders. I had to always seek out guidance from male experts, because I didn’t have access to any female mentors. It’s hard to understand that a person coming from the same background on paper doesn’t necessarily have the same opportunities or experience. This is due to inherent biases and most people propagating this don’t even realize that they are doing it. You would probably notice it only if it had happened to you. That is why having a more balanced workplace helps, because now you can have feedback from people with different experiences.
“As an engineer, I have definitely felt a lack of senior women and, as a result, lack of mentorship from female leaders.”
Starsky is doing pretty good at this front for a technical start-up. We have around 25% female engineers and women are encouraged to take on leadership tasks. As a team we realize that we still have a long way to go and are taking special efforts to reach out to female talent. With robotics specifically, it’s harder to find women because most of them are into pure software roles. There’s a trend, and it seems people follow the trend. However, more and more women are now coming into Mathematics and other fields which are more Robotics-related. So, hopefully the trend will shift soon.
What advice would you give to those who are considering a career in the autonomous industry?
When I was doing my undergrad in India, many sources of knowledge that are available now simply didn’t exist. Internet articles, YouTube lectures and online courses were not this diverse, and learning was more of a one-to-one interaction with professors and seniors than anything else. Now, the exposure is quite different and it’s more of a level playing field.
Today you don’t really have to rely on mentors to help you learn. You can find people online, do some online courses, and then do projects to put your knowledge into practice. There are many online competitions to give you examples of such projects. Start doing all of these things by yourself and eventually you will form a network that can help you grow further.
Is your team hiring now? What kind of people are you seeking?
Yes, Starsky is actively growing, and we are looking for three additional people for our team at the moment. We have immediate requirements for a Robotics Perception Engineer and a Perception Team Lead. Ideal candidates would need to have worked with Mobile Robots and done Software development. In addition, a Perception Lead should have experience working with different sensors, developing systems which use machine/deep learning and building a team.
Can you tell us any interesting facts that describe you the best?
One thing would be that I normally get things done. That is my unique selling proposition, my USP. I love working towards an answer, even when the challenge is particularly tough. Even if it’s totally unknown and I’ve never faced that problem before, I can always get to a working solution. Some people give up sooner rather than later, and I hate to give up. I like to research and try to use all my resources — my friends, my family, everybody in my network. This approach may take longer than someone who knows the solution, but I will make sure it’s done.
I frequently get assigned tasks which haven’t been started at all but need to be completed within tight deadlines. It’s a balancing act of building something which works without compromising safety or quality, and thankfully I have lots of practice doing that.
***
If you want to join the Starsky team and help us get unmanned trucks on the road, please apply here.