OnDeck AI

OnDeck is the infrastructure layer that makes Vision Language Models accessible and scalable for enterprise. We let organizations instantly find any object, behavior or event, in any footage, without needing to train a model or collect any training data.

The Pain: Creating vision models usually takes months: collecting training data, training, then deployment. Worse yet:

+ it’s often impossible to get enough data for a specific task, and

+ even the best cv models struggle to generalize across diverse camera setups, workflows and environments.

To overcome these blockers, we bet early on the power of VLMs and built a vision engine that can generalize across any task and doesn’t need any training data. We published a NeurIPS workshop paper showing our new methods with VLMs beat traditional CV even at niche tasks.

Our current customers include:

- National Defense Organizations

- Robotics Research

- Security cameras

- Behaviour analysis for port monitoring

- Off-shore oil & gas monitoring

High-Level Overview

OnDeck AI is a cutting-edge artificial intelligence company that enables analysis of any video footage without the need to train a custom model. Their technology leverages vision-language models (VLMs) to instantly identify events, behaviors, or objects in video data, eliminating the months-long process of collecting training data and model training. Originally focused on fisheries monitoring, OnDeck AI now serves a broader range of clients including governments, Indigenous nations, commercial fishers, and military organizations, providing scalable, real-time video analysis solutions that enhance monitoring, compliance, and data accessibility[1][2][4][5].

The company’s mission is to democratize advanced video analysis by making it accessible, affordable, and efficient across industries. Their AI software solves the problem of labor-intensive, slow, and costly video review by automating event detection and species identification in fisheries and beyond. OnDeck AI has demonstrated strong growth momentum, achieving six-figure annual recurring revenue (ARR) and securing pilots with high-profile clients such as the Singaporean Military, supported by notable backers like Y Combinator, National Geographic, and Eric Schmidt’s family foundation[1][4][5].

Origin Story

Founded in 2022 in Vancouver, Canada, OnDeck AI was co-founded by Alexander Dungate and Sepand Dyanatkar, childhood friends with strong backgrounds in computer science and machine learning. Sepand holds a Master’s in Machine Learning from Cambridge and has experience at the European Space Agency and Amazon, while Alexander previously built an earlier iteration of OnDeck to $150K ARR. The idea originated from the challenge of automating fisheries monitoring, where manual video review was a bottleneck. Early traction included deploying AI-powered fisheries monitoring tools on three continents and raising over $2 million in funding from environmental and tech organizations[1][2][4][5].

Core Differentiators

• No Training Data Required: OnDeck’s AI can analyze any video footage without needing labeled training data or retraining for new tasks, a major breakthrough compared to traditional computer vision models[4][5].
• Universal Species and Event Identification: Their software generalizes across various camera types, environmental conditions, and aquatic settings, enabling universal species identification and event detection in fisheries monitoring[1][3].
• Speed and Scalability: Instantly searches through terabytes of footage to find specific events or objects, dramatically reducing time and resources needed for video review[1][4].
• Strong Technical Foundation: Built on advanced vision-language models, supported by published research (NeurIPS workshop paper), and led by founders with deep ML expertise[4].
• Diverse Client Base: Serves governments, Indigenous nations, commercial fishers, NGOs, and military clients, demonstrating versatility and broad applicability[1][2][5].
• Robust Funding and Partnerships: Backed by Y Combinator, National Geographic, Schmidt Marine Technology Partners, and others, ensuring strong financial and strategic support[1][5].

Role in the Broader Tech Landscape

OnDeck AI rides the wave of AI democratization and the growing demand for automated video analytics across industries. The timing is critical as video data volumes explode and traditional manual review becomes infeasible. Their no-training-data approach leverages advances in vision-language models, a cutting-edge trend in AI that enables generalization across tasks without bespoke model training. Market forces such as increasing regulatory scrutiny in fisheries, environmental conservation efforts, and military surveillance needs create strong demand for scalable, real-time video analysis. OnDeck’s technology lowers barriers to entry for AI-powered monitoring, influencing the broader ecosystem by enabling more organizations to adopt AI solutions without heavy upfront investment in data labeling or model development[1][4][5].

Quick Take & Future Outlook

OnDeck AI is poised for significant growth as it expands beyond fisheries into broader video analytics markets, including defense and environmental monitoring. Future trends shaping their journey include continued advances in foundation models, increasing regulatory requirements for transparency and monitoring, and the proliferation of video data from autonomous systems and IoT devices. Their ability to provide instant, training-free video analysis positions them well to capture new sectors and scale rapidly. As they deepen partnerships and refine their platform, OnDeck could become a foundational infrastructure provider for enterprise video intelligence, further democratizing AI access and transforming how organizations extract insights from vast video archives[4][5].

Returning to the core value proposition: OnDeck AI fundamentally changes how video footage is analyzed by removing the need for training models, enabling instant, scalable, and affordable video understanding across diverse applications.