Lucas Young and Thomas Li, childhood friends and former software engineers at Google, have launched Deepnight, a company poised to disrupt the night vision industry. Their innovation: using artificial intelligence to transform readily available smartphone camera technology into high-performance night vision, a feat that has garnered them significant government contracts.
Traditional night vision technology, still largely analog, relies on optical lenses and chemical processes to amplify faint light. These systems, sourced from military contractors like L3Harris and Elbit America, can cost between $13,000 to $30,000 per unit. While the U.S. Army has sought to digitize this technology for years, progress has been slow.
Young, armed with a computational photography degree from Cal Poly and experience developing smartphone camera software at Google, recognized the potential of AI to overcome hardware limitations. Li, with a background in AI and computer vision, provided crucial expertise. Inspiration struck Young when he read a 2018 paper, “Learning to See in the Dark,” by Vladlen Koltun, who is now at Apple. The paper outlined using AI for low-light imaging but noted how on-device AI chips were not fast enough to support the 90 frames per second (fps) required for real-time viewing. Fast forward to 2024, and Young realized AI accelerators on system-on-chips (SoCs) had advanced far enough.
After convincing Li to join him, the pair founded Deepnight. They were admitted into the Y Combinator winter cohort. Deepnight’s approach offers a compelling alternative to the expensive hardware traditionally associated with night vision.
To gain traction with their primary customer, which was the military, Young navigated an industry event, presenting his concept as a software solution to the U.S. Army’s night vision lab. He shared a white paper that proposed this idea. A conversation with an army colonel, who was impressed by the white paper, connected the founders with people from the US Army C5ISR Center.
To demonstrate their concept’s feasibility, Deepnight developed a night vision smartphone app coupled with a smartphone-holding VR set. This primitive prototype proved sufficiently impressive, leading to their first sale. “The army awarded us a $100,000 contract in February 2024, one month into Y Combinator, based on the proof of concept in a smartphone demo and our whitepapers and presentation,” Young said.
Further demonstrations followed, including a presentation in Washington, D.C., where they showcased their software alongside state-of-the-art goggles. This presentation led to more contracts. One year after its launch, Deepnight has secured roughly $4.6 million in contracts from various federal agencies, including the U.S. Army and Air Force, and has partnered with companies such as Sionyx and SRI International.
Deepnight also quickly received investor funding. By the end of Y Combinator, the company raised a $5.5 million round led by Initialized Capital. Angel investors include Kulveer Taggar and Matthew Bellamy, the lead singer of the band Muse. Y Combinator also invested in them. Notably, Koltun, the scientist whose research initially inspired the company, became an angel investor.
Deepnight’s business model centers on software licensing and partnerships with hardware manufacturers. “Now we can make everything in the world see in the dark, because it’s just a software program. So that’s automotive, security, drones, maritime like boats, electronics, nav cameras,” Young explained. Given that their technology depends upon off-the-shelf $50 smartphone cameras, it bypasses the need for costly, custom-built hardware, offering a significant cost advantage.
