From rocket science at NASA JPL to building the AI analytics platform that enterprises actually want to use.
Most people who study physics at UC Santa Barbara and go on to work at NASA's Jet Propulsion Laboratory do not end up building AI-powered analytics platforms. Jon Koury did. That career pivot from the physical world of cryogenics and aerospace to the digital infrastructure of enterprise data is not a contradiction - it is a through-line. Systems thinking, at whatever scale, is what Koury does.
Based in Huntington Beach, California, Koury joined Omni Analytics as a founder at a moment when the business intelligence market was about to be rewritten. The company, launched in 2022 out of San Francisco, had a simple premise: the tools companies used to understand their own data were broken. Too technical for most users, too slow for anyone, and too disconnected from the AI capabilities the world had just discovered.
"AI isn't replacing analytics - it's expanding it."
- Omni AnalyticsOmni's approach sits at the intersection of two ideas that previously competed: a governed semantic layer (which keeps data consistent, permissioned, and reliable across an organization) and conversational AI (which lets non-technical users ask questions in plain language). The architecture Koury and the team built does not force a choice between those two things. It runs both at once.
The market responded. By April 2026, Omni had closed a $120 million Series C led by ICONIQ, with participation from GV, Theory Ventures, First Round Capital, and Redpoint Ventures - valuing the company at $1.5 billion. Revenue tripled in the run-up to the round. Customers consolidating legacy BI tools onto Omni included BambooHR, Checkr, Cribl, dbt Labs, Guitar Center, Mercury, Pendo, and Synthesia - an unusual mix of HR software, financial infrastructure, developer tools, and retail that speaks to how broadly the data problem extends.
Koury's path to the analytics world ran through the physical sciences. After graduating from UC Santa Barbara with a physics degree around 2020, he worked at NASA's Jet Propulsion Laboratory - the same facility responsible for Mars rovers and deep space missions - and then at Cryo Innovations, a company working in cryogenic technology. These are not typical stepping stones into SaaS, which is precisely what makes his trajectory interesting.
The shift toward automation and AI systems architecture came next. Before joining Omni, Koury worked as an Automation and AI Systems Architect - designing the kind of integrated, automated workflows that modern companies depend on but rarely think about until they stop working. That specialization in systems, in how pieces connect and what happens when the connections fail, is visible in Omni's product philosophy.
Omni's semantic layer is not just a feature. It is the structural bet the company made: that data, to be useful at scale, needs a single source of governed truth from which all analyses flow. SQL analysts, business users, and AI models all query the same underlying model. The chaos that typically emerges when different teams define "revenue" differently is exactly the problem this architecture prevents.
Omni's culture is worth noting because it reflects what Koury signed up for. The company posts weekly engineering demos on YouTube, a practice that signals something about how they think about accountability and velocity. Performance is not a background concern - it is a weekly conversation. Engineers at Omni are asked to quantify how changes affect query speed. The demos exist to make that expectation visible.
With 770-plus team members spread across San Francisco, Santa Cruz, Philadelphia, Toronto, Dublin, and Sydney, Omni has grown from a thesis about what analytics should feel like into a global operation. That growth - which brought the company from its initial funding to a $1.5 billion valuation in under four years - reflects both the scale of the problem and the quality of execution required to tackle it.
The Series C included a $30 million employee tender offer, a detail that matters because it signals that Omni's investors are treating this as a long-duration bet, not an exit sprint. ICONIQ's Matt Jacobson framed the opportunity clearly: the barrier in data has shifted from access to understanding. Providing the understanding is where Omni operates - and where Jon Koury works every day.
"The barrier in data has shifted from access to understanding."
- Matt Jacobson, Partner, ICONIQWhat Omni's customers are actually buying is not a dashboard tool. They are buying the ability to trust their numbers - and to get answers to questions their data team was never big enough to handle. AI-powered natural language querying, role-based permissions, row- and column-level security, embedded analytics for external-facing products, and real-time collaboration across an enterprise: Omni bundles these not as features but as a coherent architecture for how data should move through an organization.
Koury's Huntington Beach home base and Omni's San Francisco headquarters capture something about how this generation of startups operates - distributed by default, connected by product conviction. The physics graduate who once worked on cryogenic systems and space-age hardware now works on data infrastructure. The intellectual restlessness is the constant. The domain changes; the systems thinking stays.
In 2026, enterprise data is at an inflection point that happens to be exactly where Omni is positioned. Companies that once ran on spreadsheets moved to data warehouses. Data warehouses produced enormous volumes of queryable information that most employees could not access without a data analyst intermediary. AI changed the query problem. But it introduced a new one: which AI-generated answer should you trust, and why?
Omni's semantic layer answers that question structurally. The governed model defines the rules. The conversational AI runs within those rules. The result is a system where a non-technical user can ask a business question in plain English and receive a reliable answer - not because the AI is smarter, but because the architecture beneath it is designed to catch errors before they propagate.
This is the problem Jon Koury helped build a solution for. Whether the trajectory continues from $1.5 billion upward is not guaranteed - nothing in software is - but the infrastructure Omni has put in place is the kind that takes years to replicate. In a market measured in days and weeks of improvement, that is a durable advantage.
Omni's architecture rests on three interlocking ideas. Get one wrong and the whole thing collapses. Get all three right and you have the first BI platform that actually works for everyone in an organization.
One shared definition of truth. Metrics, calculations, and access rules defined once and applied everywhere - so "revenue" means the same thing across finance, sales, and engineering.
Natural language querying that runs within the governed model. Non-technical users get reliable answers without needing a data analyst as an intermediary.
Row- and column-level permissions, role-based access, multi-tenant architecture, and SOC 2 compliance built into the foundation - not bolted on afterward.
Bi-directional dbt integration that lets data teams work in the tools they already use while Omni serves as the analytics layer on top.
Customer-facing dashboards and reports built with Omni's infrastructure, so product companies can ship data experiences without building separate BI tooling.
Performance is a weekly conversation at Omni. Engineering demos measure it. Customers depend on it. Slow BI tools get abandoned; fast ones get used daily.
Omni went from founding to unicorn status in under four years. Revenue tripled in the year leading to the Series C. The round, closed April 2026, included a $30M employee tender offer - a signal that this is a long-duration play, not an exit sprint.
The lead investor, ICONIQ, noted that the market opportunity in data "is just enormous" - bigger than business intelligence as traditionally understood.
Omni's stack reflects a bet on modern cloud-native tooling. TypeScript, React, and Remix on the frontend. AWS infrastructure throughout. AI models from Anthropic and OpenAI powering natural language features. dbt at the data layer.
From HR software to fintech to music retail - Omni's customer list spans industries precisely because the underlying problem (unreliable, inaccessible data) is universal.