Skip to the stories
Windows & PowerShell

Disseqt AI Raises $8m to Test AI Agents Before They Go Live

Dublin-based Disseqt AI has raised $8m to scale its pre-launch testing and runtime control platform for enterprise AI agents, led by Gresham House Ventures.

Disseqt AI Raises $8m to Test AI Agents Before They Go Live

Disseqt AI has raised $8m, and the pitch is aimed squarely at the part of enterprise AI that's stuck. Plenty of organisations can build an agent. Far fewer can show that it behaves safely, fairly and within policy before it touches customers or money. The Dublin company says the new funding will go towards deploying and scaling AI agents in line with regulatory requirements, and it's a useful signal of where enterprise spending is heading.

Two South Asian men and a white man in business casual clothing against an AI generated office background.

Irish co-founded Pretxt raises $5m seed to supervise AI agents

From left: Manish Atri, Apoorva Kumar and Cyril Treacy. Image: Disseqt AI – background AI generated.

Irish-founded End Close raises $3.5m in seed funding round

What the round is meant to fund

The money has two jobs. The first is international expansion. Disseqt is headquartered in Dublin, with additional locations in London and Bengaluru, and the company is looking to take its platform to more enterprises. The second is hiring. Disseqt intends to recruit across sales, operations and technology, which is the usual shape of a young software company moving from a founding team into a go-to-market operation.

Germany’s Agile Robots raises $500m at $8.5bn valuation

The company was established in 2025 and is headquartered at NovaUCD, the University College Dublin innovation hub. Its founding leadership is Apoorva Kumar as CEO, Manish Atri as CTO and Cyril Treacy as COO. The stated mission is to help companies test AI applications and agents before they launch.

Bloomberg: Anthropic IPO six-times oversubscribed

Who's backing it

The round was led by Gresham House Ventures, with participation from Sure Valley Ventures, Delta Partners and Enterprise Ireland. Angel investors include senior leaders from OpenAI, Palo Alto Networks, Salesforce and Meta, plus former cricketer Rahul Dravid.

UK AI chips start-up Fractile raises $720m at $7bn valuation

The angel list is worth a second look. Their participation doesn't tell you the product works, but it does suggest the founders are talking to the right people about the right gap.

Dublin’s Akara raises $5.5m to accelerate hospital AI tech

The pitch: test before launch, control in real time

Kumar framed the problem bluntly. "Every enterprise is handing real work to AI agents, but very few can prove those agents are safe, fair and within policy," he said. "That gap is what stops AI from going live, and it is what we built Disseqt to close." According to Kumar, the platform tests before launch, controls behaviour in real time and keeps an audit trail, independently of the model providers.

That last point matters more than it might look. Many teams today rely on whatever safety tooling their model vendor ships with. A separate assurance layer, which sits outside the model and records what the agent actually did, is a different proposition. It gives an IT or risk team evidence that doesn't depend on the vendor's own account of its behaviour, and it can apply the same rules across more than one model.

Kumar also set out the stakes. "When your agent is talking to 80m people, or processing thousands of banking transactions, 'good enough' isn't an option," he said. "This funding lets us bring that standard to many more enterprises. AI governance and assurance needs to become a standard part of how every enterprise runs AI." Those figures are the company's own illustrations of scale, not measured results from a deployment.

Recent recognition

In September 2026, Disseqt won the Irish leg of the 2026 KPMG Global Tech Innovator competition, which earned it a place in the global finale. That finale took place in Lisbon, Portugal, in early November. Competition placings are useful as a credibility marker with enterprise buyers, though they don't substitute for reference customers or independent evaluation of the product itself.

Why this matters for IT teams

My read is that the company is betting on a shift in how enterprises buy AI. The first wave of agent projects was mostly about capability: can the agent do the task? The next question is operational: can you prove what it did, stop it from doing the wrong thing, and show an auditor the record? Teams in this position typically find that the testing and monitoring work takes longer than the build, and it often falls to a small platform or security group with no dedicated tooling.

If you're running or planning agent workloads, a few questions are worth putting to any vendor in this space, Disseqt included. Does the assurance layer work across the models you use, or only one? Where do the audit logs live, and who can read them? How does pre-launch testing map to the policies your business already has, and who maintains those rules once the agent is live? Those answers will tell you more than the funding headline.

Disseqt's $8m round is modest by the standards of frontier AI, but the company is aiming at a problem that most enterprises will hit as agents move from pilots into production. Watch for the hiring and international expansion to show whether the team can turn a strong founding pitch into repeatable deployments. That's the point where the claim that AI governance should be a standard part of enterprise operations will either hold up or stay a slide.

Source: Laura Varley · www.siliconrepublic.com

Same flag