Who we are

Latent AI is a well-funded spinout from SRI International, based on years of DARPA research.  We design advanced, yet easy to use, software developer tools that take the hard work out of embedding complex AI models into lean edge devices.  We transform the way developers train AI for the edge, enabling rapid AI deployment, speed to market and scalability.  Our patented Adaptive AI™ technology can be used on any edge hardware, framework, OS or architecture -- and gives us a true competitive market advantage. We are on contract with various services within DoD and are on track to achieve 6x growth this year. 

Latent AI is backed by Future Ventures, Blackhorn Ventures, Auto Tech Ventures, Booz Allen Hamilton,  Lockheed Martin, Standard Investments, Pegasus Tech Ventures, and SRI International.

What we do

Latent AI enables creation of more robust, efficient, and agile edge applications that improve user adoption and mission impact.Our current product mission is simplifying and accelerating traditional AI workflows from model training and optimization to deployment and support. Our flagship product, the Latent AI Efficient Inference Platform™ (LEIP), is a unified SDK built for machine learning engineers and AI software developers. Our solutions are being used in more places than ever before including manufacturing automation, predictive maintenance, process control, consumer electronics, and defense and aerospace applications. And we are just getting started.

Who we're looking for

We’re looking for an experienced, hands-on ML/AI engineer (typically 2-3 years of relevant experience, or a PhD with equivalent expertise) to join our team building the next generation of tooling for ML/AI on the edge.

What you’ll do

• Balance independence and collaboration: you make quick, informed decisions on your own, but stay aligned with the team and company goals.
• Value both agency and teamwork, and bring strong communication skills to keep collaboration productive.
• Enjoy supporting the growth of others: sharing knowledge, mentoring junior teammates, and helping the whole team level up.
• Extend our technology and products for edge MLOps: clean interfaces, reliable services, and ergonomics that make ML workflows fast and repeatable.
• Build products that enable users to explore, assess, and improve training data quality at scale (e.g. embeddings, clustering, distillation).
• Enhance automated MLOps pipelines with new models, optimization strategies, and evaluation methods to mine predictive signal from large experiment corpora.
• Help evolve our agentic layer so customers can interact with the system naturally (query state, orchestrate runs, interpret results).
• Ship production-grade code with strong CI/CD, tests, and docs; raise reliability and developer experience across the stack.

How we decide

An ideal candidate will be someone with;

• Technical depth in ML/AI with a preference for computer vision and NLP
• Proven contributions, whether in industry or academia, that moved innovative ideas closer to product. 2-3 years industry experience
• Hands-on execution: the ability to design, build, and ship, not just theorize.
• Collaborative independence: you’re comfortable making autonomous progress, but you thrive in a team environment.
• Growth mindset: curiosity, adaptability, and willingness to learn new approaches and tools.
• Are fluent in Python and modern ML frameworks like PyTorch; familiarity with agentic AI frameworks is a bonus.
• Know your way around CI/CD, automated testing, and documentation, and care about building robust systems.
• Have some familiarity with data science front-end/UI workflows to support quick demonstrations (streamlit, dash, gradio).
• Have a track record of driving innovation from research or prototypes into production ready features or products
• US citizenship preferred but not a requirement
• Candidates based in NJ preferred, remote/hybrid possible