Women in Robotics Bay Area lead shares lessons on autonomy

Sanjuksha Nirgude has spent nearly a decade working across the broader robotics landscape — from warehouse mobile robots to space simulation with NASA’s VIPER lunar rover to ground robotics. Today, she serves as an Autonomy Technical Lead at Nightingale Security, where she designs and deploys autonomous aerial systems for 24/7 security and critical infrastructure operations.

She holds a Master’s degree in Robotics Engineering from Worcester Polytechnic Institute (WPI).

In addition to bridging advanced robotics research with real-world production systems, she serves as an organizer for the Women in Robotics Bay Area chapter, advocating for gender diversity and mentorship in autonomous technology. 

And, she found a moment in her busy life to chat with The Drone Girl! Here’s our conversation, lightly edited for clarity and length.

Do you know an awesome drone girl I should profile? Contact me here.

The Drone Girl: What’s the top challenge within drones and robotics that you’re trying to solve for right now?

Sanjuksha Nirgude: Multi-drone coordination. That’s not just getting a fleet to safely share the same airspace without colliding, but pushing into the many applications that open up once multiple drones are actually working together as a system rather than as individual vehicles doing their own thing.

TDG: Ah ha, this has me thinking of drone light shows, which are such a good example of that! Now one thing on my mind lately is how AI fits into this.

SN: Yes, so the next frontier I’m excited about is bringing physical AI into that picture, specifically, how vision-language-action models, which have mostly proven themselves on ground robots and manipulators, can translate into aerial robotics. Drones operate in a much less structured, faster-moving environment, so figuring out what it takes for a VLA-style model to reason and act in that context feels like one of the more interesting open problems in the field right now.

TDG: Okay, so I’m curious how your background — which spans warehouse logistics, space simulation, and ground robotics all leads into this. How has working across ground and space systems shaped your approach to drone autonomy software?

SN: What’s stuck with me is how much of the core framework actually carries over. The underlying principles of robotics, how you think about perception, planning, control, vision, how you debug a system in production, how you handle a customer issue or ship a new feature or product from scratch, all of this translate more smoothly from ground and warehouse robotics into aerial than people might expect.

ROS(Robot Operating System) is a good example. It has been the backbone across nearly every domain I have worked in, from lunar rover simulation to warehouse AMRs to drones now. The vehicle underneath looks completely different in each case, but a surprising amount of the underlying approach carries over regardless.

TDG: What about the differences?

SN: There is something about working on robots that fly that simply hits differently, and my excitement for this field increased tenfold once the systems I was building left the ground. But the more interesting realization, technically, is just how much research and knowledge from ground robotics turns out to be transferable to drones. You are not starting from zero every time you move domains. You are building on the same foundation and applying it somewhere new.

TDG: Now you’ve worked heavily on high-fidelity simulation environments. from NASA lunar rovers to autonomous warehouse bots. When taking a drone out of simulation and deploying it into the real world for 24/7 security operations, what is the ‘gap’ between code and reality that surprises engineers the most?

SN: Closing the sim-to-real gap is a huge area of research right now, especially with the rise of world models, where physical AI and robots are trained inside a simulated world with the goal of that training holding up once the robot is operating in the real one. Most of that effort has gone into getting the physics right, the visuals right, how realistic and fast objects move, frame rates, even weather conditions. And simulation has genuinely gotten good at all of that.

TDG: Any surprises for you there?

SN: Network connectivity is one of the biggest gaps that gets the least attention. Ground robots are usually operating in an enclosed space or on a stable wired or WiFi connection.

Drones don’t have that luxury. When you’re flying miles from a ground station, the strength and consistency of that connection has a direct impact on whether operations run smoothly. Simulating a realistic network environment, with latency, dropout, degraded signal, is still a much smaller focus than simulating physics or weather, even though it’s one of the most common real-world failure points. That gap matters whether you’re talking about a rover on the Moon or a drone running security operations around the clock.

TDG: What’s something you think people don’t understand about your work in drones and robotics?

SN: That it’s nowhere near as software-centric as people assume. Robotics is genuinely one of the most interdisciplinary fields in engineering, you need mechanical, electrical, and computer science all working together, and that’s still true whether you’re building a ground robot or a drone.

Up until fairly recently, very few universities even offered a dedicated robotics engineering degree, so most people in the field came in through one of those three lanes and had to pick up the other two along the way. I started in mechanical engineering myself, and I think that foundation still shapes how I approach software design today, in the sense that I’m always thinking about it in the context of the robot’s anatomy, not just as code sitting on top of a machine.

TDG: Is there one area people should focus on though if they want a career like yours?

SN: The coding and AI side gets most of the spotlight because it’s the most visible part, but the algorithm is only ever as good as your understanding of the system underneath it. A working knowledge of robot design, hardware behavior, and sensors, how they’re placed, calibrated, and how they perform under real-world conditions, directly shapes how well you can write that algorithm in the first place.

A plan that looks perfect on paper can fail entirely if it doesn’t account for what the platform underneath it can physically do. That intersection, not any one piece of it, is what makes robotics hard and honestly what makes it interesting.

TDG: We were talking previously, and you mentioned no two days in your life look the same.

SN: That’s genuinely one of the top reasons I love my job. Robotics is such an interdisciplinary field that being effective means being comfortable moving across a wide range of problems, not staying confined to one lane. That range is something I’ve come to see as a strength rather than a distraction.

TDG: Okay, well do you have any average day in your life? Or one day that’s particularly stood out?

SN: When I joined a site assessment to evaluate whether a location we were installing at was actually suitable for the application. Most of my day-to-day work is on the software side, so being out there in person, thinking through airspace considerations and hearing directly from the customer what they needed the system to do, gave me a level of understanding you simply don’t get from a behavior planner or a simulation alone. It’s the kind of context that makes the software work better once you’re back at your desk.

TDG: Ah ha, I agree! The days out in the field are way better than my days at my desk.

SN: Both are core to the job. One day I’m deep in the literature shaping how we think about a problem, the next I’m on-site seeing exactly what the system needs to deliver in practice. Staying sharp on both ends, the research and the real world, is what it actually takes to build something that holds up.

TDG: Fair advice! Okay, so for the women and aspiring engineers reading The Drone Girl who want to break into autonomy software: What is one core technical skill or mindset shift you recommend focusing on first?

SN: Speak up and ask questions, even when you’re not sure you have the “right” answer yet. I’ve noticed women in particular tend to hold back, assuming their input isn’t important enough to share. But you never actually know until you put it out there, and often it’s the question everyone else overlooked that ends up moving a project forward. Getting out to local events and talking directly to people already in the field is one of the best ways to build that instinct, you learn faster from real conversations than from any course or textbook.

The second one is more specific to robotics: I’ve learned to keep asking myself, how can I simplify this? Robotics gives you access to genuinely sophisticated algorithms, sensors, and compute, and it’s tempting to reach for the fanciest tool just because it exists. But the best engineering solution is usually the simplest one that reliably gets the job done. Knowing when you don’t need something complicated is just as important as knowing how to build it.

TDG: As an organizer for Women in Robotics Bay Area, what are some of the most pressing hurdles you see early-career women facing in autonomous systems today?

SN: Confidence is a big one. I’ve seen women who are genuinely excellent at what they do still doubt themselves simply because the field feels daunting, or because they don’t see many people who look like them further along the path.

Also feeling like you need to prove you belong in the field because you’re a woman. That’s extra weight that shouldn’t be there in the first place.

The truth is, robotics rewards curiosity, creativity and skill, and there’s room to build a career for anyone willing to work hard in any corner of it, hardware, software, systems, whatever draws you in. The sooner that pressure to prove yourself goes away, the more women will simply see themselves as engineers, not as an exception in the room.

TDG: And what kinds of things could we do better at (or is Women in Robotics Bay Area doing) to support that?

SN: A big part of it comes down to visibility, exposure to the technology itself, but just as importantly, exposure to the people, especially women, already doing incredible work in this industry. Seeing someone who’s already achieved what you’re working toward is one of the most powerful things for an early-career woman to witness. It turns something abstract into something concrete: this is possible, and here’s someone who’s done it.

That’s a big part of what Women in Robotics focuses on. We regularly organize networking events and talks featuring women from both industry and academia, giving them the spotlight and letting their work speak for itself, similar to what The Drone Girl is doing with this very profile series.

The other piece is mentorship and community…making real openings for women to ask questions of people more senior than them, or even peers just a step or two ahead, about how they actually got into the field or how to navigate it. A lot of that knowledge doesn’t live in a textbook, it lives in conversation, and it tends to happen naturally once you’ve built an actual community around it. That’s what we try to create! It’s a space where those conversations keep happening, where mentorship is simply part of how the community works.

Follow Sanjuksha Nirgude on LinkedIn, and follow along with Women in Robotics Bay Area on Meetup.

Do you know an awesome drone girl I should profile? Contact me here.

Related