Self-driving cars are a dream decades in the making, but now that they’re finally here, manufacturers have to deal with the harsh reality that making them safe to use is monumentally harder than it sounds.
When vehicles have to make life-or-death decisions without human input, the engineering behind autonomous vehicles becomes crucial. This is the work of systems engineer Amardeep Sidhu, who leads the system safety team at a leading automotive technology provider. As he refines the tech that empowers AI to make better decisions in complex settings, his work is a constant tightrope walk of automotive innovation and functional safety.
Here’s how Sidhu is working to make autonomous driving both viable and trustworthy.
Building a Career in Complex Systems Safety
Amardeep Sidhu has long held “a deep curiosity about how things work — and, more importantly, how they can work better.” It’s a passion that led him to study at Purdue University and MIT, then dedicate his career to complex systems design.
For much of his professional life, he’s been motivated by a pivotal moment that came during his graduate studies, while he was working on diagnosing problems in lithium-ion batteries. “I realized that even small improvements in how we detect faults or design safety systems could have a huge impact on people’s lives,” he recalls. This realization paved the way for Sidhu to see engineering not just as a technical field but as a way to contribute to safer, more sustainable technologies for the future.
Today, he’s a lead systems engineer at a leading automotive technology provider. There, he manages a multi-continent team and oversees the development of safety systems for technologies like lane-keeping assistance, emergency braking, and other automated and autonomous driving features.
In Sidhu’s opinion, true autonomy is no longer a technology problem; it’s a systems problem. In order to ensure maximum reliability, developing autonomous vehicles requires a holistic approach to the development of sensors, control algorithms, and human-machine interfaces — one that embeds safety considerations from the earliest stages of design instead of relegating it to an afterthought.
Redefining Safety for AI-Driven Systems
While it’s easy to tell a self-driving car that it must stop at a stop sign or slow down when the speed limit changes, the reality of developing systems for safe autonomous vehicles is infinitely more complex than you might think.
What happens when there’s road work that does not match the training data in the car’s internal memory? What if it needs to take a path that’s not on any GPS? How does the car’s computer decide when to surrender control to a human driver?
These complex situations cannot be easily answered with black-and-white if-then statements. They require the creation of robust, dynamic systems that adequately prepare the vehicle for any scenario.
Much of Sidhu’s work has been devoted to creating and refining these systems. By equipping a vehicle with a litany of sensors, detailed digital maps, and a control brain that synthesizes data from both sources and makes real-time decisions, self-driving cars can adequately and reliably assess safety conditions and activate driver assistance features if necessary. In fact, you may have already seen them in action — products featuring Sidhu’s work on systems like these have been deployed on public roads since 2022.
Sidhu is also exploring ways to leverage AI to improve safety processes themselves. Currently, he’s investigating how large language models can support safety documentation and validation, potentially enabling faster, more traceable development cycles while maintaining the rigorous standards required for safety-critical applications.
Setting Standards and Shaping the Industry
As the automotive industry increasingly relies on machine learning, Sidhu believes in the necessity for clear guidelines and industry standards. To this end, he contributed to SAE J3187 and SAE J3307, standards that guide best practices in automotive safety and systems engineering.
“These aren’t just technical checklists,” Sidhu explains. “They’re shared languages that allow manufacturers, suppliers, and regulators to coordinate efforts around increasingly complex technologies.”
Through these standards, Sidhu is helping define how the industry measures safety in automated and autonomous systems.
Sidhu’s impact on the industry extends beyond these technical guidelines. Since 2019, he’s served as a subject matter expert and teaching assistant for MIT’s professional education programs in systems engineering. These courses have reached over 7,000 professionals from more than 1,000 organizations, including Boeing, General Motors, and the U.S. Navy.
“My role involves helping mid-career engineers understand and apply modern systems thinking and decision-making frameworks in their day-to-day work,” Sidhu explains. “What makes this work deeply meaningful is the ripple effect: these professionals take what they learn and apply it to real-world challenges across sectors like Aerospace, Automotive, Tech, and Healthcare.”
Bridging Innovation and Responsibility
As the autonomous vehicle industry charges forward, it must always seek to balance technological advancement with uncompromising safety. In Sidhu’s opinion, these goals aren’t mutually exclusive — they’re interdependent.
His contributions are already in action and on the road today. His work on industry standards has helped create common frameworks for automotive safety. And through his academia efforts, he’s equipped thousands of engineers with better approaches to systems safety.
The methodologies refined by engineers like Amardeep Sidhu will only become more important in the coming years — not just for the automotive industry, but for any field where AI systems make consequential decisions. The future of safe autonomy depends on this careful balance of innovation and responsibility.
