A week of bewildering infrastructure failures is a nasty time to discover nobody can explain the machinery. That was the backdrop for Atalanta CEO and co-founder Anjana Rajan at the National Academies of Sciences, Engineering, and Medicine's Cyber Resilience Forum, which concluded November 20 after two days in Washington. Her central point deserves more attention than the usual cybersecurity conference fog: operators often do not sufficiently understand how their software-heavy systems will act when conditions turn abnormal or openly hostile.
Rajan calls it the software understanding gap. The phrase is plain, almost annoyingly plain, and that is why it lands. Transportation, banking, telecoms, national security, energy—these are not neat, self-contained applications. They are sprawling arrangements built, patched, maintained, and operated by different parties over years. When one piece coughs, the people responsible can be left squinting at dashboards and guessing. That is no way to run critical infrastructure.
Math Is Being Pushed Closer to the Operator
Atalanta used the forum to announce Ata, a new formal verification language. The stated aim is to put mathematical verification in the hands of operators so they can test how critical systems should behave under normal, abnormal, or hostile conditions. Rajan's presentation, titled Forward Deployed Formal Methods, focused on translating mathematical rigor into real-world missions rather than leaving it trapped in research papers and specialist teams.
That distinction matters. Formal methods have long carried a reputation for being powerful but difficult, slow, and confined to high-assurance corners of engineering. If Ata genuinely makes verification more usable for the people running consequential systems, that could address a very real pain point. If it merely creates another layer that requires rare experts to interpret, then the practical impact gets narrower in a hurry. Investors watching this field should not confuse a compelling problem statement with frictionless deployment. Those are two very different animals.
“How do I know my critical infrastructure will perform as intended, even when under attack?” Rajan asked. That is the operational question underneath all the jargon.
Demand Is Obvious; Adoption Is the Hard Part
The forum itself was not a small gathering of software hobbyists. Leaders from the NSA, the Department of War, AWS, Google, Microsoft, and other organizations discussed the technical and operational challenge. That lineup signals that software assurance has moved beyond a niche concern, especially after major outages exposed vulnerabilities across essential services. Nobody needs to invent urgency here. It is already sitting on the table, glowering.
Still, critical-infrastructure buying cycles can be punishingly slow. Security teams, engineers, mission owners, regulators, procurement offices, and legacy vendors all get a say. An elegant tool must fit existing workflows, prove it can handle messy systems, and earn trust from people who cannot afford an outage caused by the very solution meant to prevent one. The sales pitch may be straightforward; the operational proof is where the bruises happen.
Existing Work Gives the Story Some Weight
Atalanta is not presenting Ata in a vacuum. The company says it is working across defense, space, and energy. In August, Viasat announced it had used Atalanta's Argo platform to demonstrate improved satellite-network resilience against modern warfare threats. Atalanta is also supporting the Department of Energy's Genesis Mission through a partnership with Idaho National Laboratory, applying formal verification techniques to autonomous nuclear-reactor development.
Those are meaningful reference points because satellite communications and nuclear systems are exactly where vague promises should be treated with suspicion. The stakes are high, the engineering is unforgiving, and verification is more than a fashionable word. But the public details remain limited. We do not have contract values, revenue figures, deployment scale, renewal evidence, or independent performance measurements for Ata. Anyone pretending this announcement alone settles the commercial case is selling smoke.
What Sector Watchers Should Keep an Eye On
- Whether Ata moves from a launch announcement into named operational deployments.
- Whether Atalanta can show that operators—not only formal-methods specialists—can use the language effectively.
- Whether work with Viasat, Idaho National Laboratory, defense, space, and energy converts into repeatable programs.
- Whether the recent outage anxiety produces budgets for prevention rather than another round of reports after the fact.
My read is cautiously positive, with the emphasis on cautiously. Rajan has identified a problem that is getting harder to wave away: complex software systems can fail in ways their owners cannot readily explain. Atalanta's bet is that mathematical verification can give mission owners a better answer before the lights flicker. The need is credible. The company now has to prove that the method can travel from high-stakes demonstrations into durable, usable practice. That is the whole ballgame.