Yarden Gross, CEO and Cofounder of Orca AI – Interview Series

Yarden Gross is an experienced entrepreneur with vast knowledge and experience in maritime technologies.
He currently spearheads Orca AI’s vision to make autonomous commercial shipping a reality. Before founding Orca AI, he was the co-founder and CEO of Engie, a VC-backed automotive tech company. He holds a BA in Economics and Business from Reichman University (IDC Herzliya).

Orca AI is a maritime technology company that uses computer vision and artificial intelligence to help ships navigate more safely and efficiently. Its platform fuses data from cameras, radar, and AIS to create continuous real-time awareness, reduce collision risk, lower fuel use, and ease crew workload. Products like SeaPod, FleetView, and Co-Captain support vessel monitoring, fleet oversight, and shared situational data, enabling a step toward autonomous shipping. The system is deployed across global fleets and powered by one of the world’s largest maritime visual datasets.

You’ve now spent over seven years building Orca AI, after previously founding companies in automotive diagnostics and repair. What originally motivated you to shift from land-based mobility tech into the maritime world, and what problem did you set out to solve when you launched the company?

I’ve always been driven by a desire to solve complex problems using technology that makes a tangible difference. My background in automotive diagnostics exposed me to the power of data and real-time decision-making. When I transitioned to the maritime sector, I saw an industry ripe for innovation. Traditional maritime navigation methods were heavily reliant on manual processes and outdated systems. Growing up on the shores of the Sea of Galilee, I developed a deep respect for the challenges of navigation. Orca AI was founded to bring the power of AI and computer vision to improve safety, reduce human error, and enhance operational efficiency at sea – addressing gaps in the industry and unlocking the full potential of maritime data.

Orca AI was founded at a time when the maritime sector was still heavily reliant on legacy navigation practices. What gaps did you observe early on that made you confident AI and computer vision could meaningfully improve safety at sea?

The maritime sector was facing significant challenges: over-reliance on radar and AIS for navigation, limited integration of modern sensors, and a lack of actionable insights from the data being gathered. I recognized that while these legacy systems were useful, the level of real-time, intelligent decision support needed to manage modern maritime risks effectively was missing.

By integrating AI and computer vision, we could transform raw data into actionable intelligence, enabling crews to not just react, but anticipate and prevent potential risks. That’s where AI’s true value could be unlocked, enhancing safety, operational efficiency, and situational awareness.

Co-Captain has been described as a “Waze of the Seas.” What were the biggest technical challenges in building a real-time platform capable of interpreting sensor feeds, vessel behavior, and environmental risk at global scale?

The biggest challenge was ensuring that Orca AI’s systems could process vast amounts of data from diverse sources, such as AIS, radar, and cameras, and make sense of it in real time. This required advanced algorithms capable of interpreting complex sensor feeds and understanding the behavior of vessels in different environmental conditions. Achieving global scale meant dealing with a variety of geographical, weather, and regulatory environments, all of which required us to build a robust platform capable of learning and adapting to these nuances. Building a system that could work across diverse shipping lanes, integrating all the information in a seamless way, was no easy feat.

Maritime environments present edge cases that are far more unpredictable than roads—fog, glare, rogue waves, unusual vessel types, and piracy zones. How did you train your models to operate reliably in such conditions?

Training the AI to handle edge cases required a mix of real-world data collection and simulation. We worked closely with shipping companies to gather real-world data from challenging environments, ensuring our models could handle the unpredictability of maritime conditions. We then used these data sets to train the AI, simulating extreme weather conditions and rare events to ensure our system could adapt in real time. It’s a continuous learning process, where the models are constantly trained and retrained based on new data to improve their reliability in challenging conditions.

Orca AI works in some of the most congested shipping lanes on the planet. What breakthroughs in perception, detection, or fusion allowed you to move from traditional alert systems to true situational awareness?

The breakthrough came not from adding more data, but from making existing data smarter and more actionable. Traditional alert systems simply notified the crew when a potential threat was detected. We’ve taken it a step further by combining radar, AIS, and visual data from our SeaPod units. By fusing these data sources, we’ve been able to eliminate irrelevant signals, reduce noise, and create a clearer, more accurate picture of what’s around the vessel. This intelligent fusion allows our system to provide context – such as how nearby vessels are behaving or whether a situation may escalate – so the crew can make informed, proactive decisions.

The ability to detect unusual vessel behavior is becoming increasingly important. How is AI reshaping the way fleets identify risks such as erratic navigation, collisions, or potential piracy?

AI allows us to identify deviations from normal behavior earlier than traditional systems. Instead of waiting for a risk, like a collision or piracy threat, to fully develop, Orca AI continuously analyzes vessel movements, speed, and surrounding conditions. By monitoring these patterns in real time, the system can flag early signs of potential risk – like erratic navigation or unusual behavior – which gives crews the time they need to act. This shift to proactive risk management is key to transforming maritime safety and operations.

Orca AI’s ‘Co-Captain’ enables vessels to share alerts with one another in real time. What does this signal about the future of collaborative maritime intelligence networks?

By enabling vessels to share data and alerts in real time, we’re creating a network where ships can learn from one another and make more informed decisions. This will lead to a shift from isolated decision-making to a more connected, cooperative approach. Over time, these networks could expand regionally or fleet-wide, with the ultimate goal of providing clearer, faster, and shared decision support across the maritime ecosystem. It’s about creating a more intelligent, interconnected maritime environment where risks can be anticipated, not just reacted to.

Your recent $72.5M raise marked the largest funding round in maritime technology to date. How does this level of investment change your roadmap, especially as the industry begins accelerating toward autonomous shipping?

The funding accelerates our mission, enabling us to expand and scale faster. It doesn’t change our core roadmap, which is focused on intelligent decision support, but it will allow us to invest more in R&D, data acquisition, and strategic partnerships. As the industry moves toward autonomy, this investment helps us refine our platform to provide the real-time, reliable data needed for autonomous systems to thrive. It strengthens our commitment to supporting human decision-making in the near term, while also preparing the industry for autonomous vessels in the future.

As fleets look to reduce emissions and improve operational efficiency, where do you see AI having the biggest near-term impact beyond navigation and safety?

Beyond navigation and safety, AI can significantly impact operational efficiency in areas like predictive maintenance, fuel optimization, and reducing emissions. AI can analyze real-time performance data to predict maintenance needs before they become issues, ensuring assets are utilized more effectively. It can also provide insights on fuel consumption patterns, helping vessels optimize fuel use and reduce emissions. The key is using AI to provide actionable insights that enable smarter decision-making, which ultimately drives efficiency and sustainability.

Looking ahead five years, what role do you believe AI-powered situational awareness will play in moving the industry closer to autonomous or semi-autonomous vessels, and what milestones must be achieved to get there?

AI-powered situational awareness will be essential for autonomy, but the biggest hurdles ahead aren’t technological – they’re legal and regulatory. The challenge isn’t whether AI can detect risks; it already does that well. The real challenge is creating the legal framework that clarifies liability when AI supports decision-making, and ensuring regulations evolve to govern AI use effectively. Until that framework is in place, the human remains in command. Our goal is to continue strengthening human decision-making with AI support, so that the industry can safely transition to autonomy when the time comes.

Thank you for the great interview, readers who wish to learn more should visit Orca AI.