- Tangos has raised a $20 million seed round led by Red Dot Capital Partners, with participation from Leaders VC, SignalFire, Venture Israel, Clarim, Clutch Capital, Selah Ventures and a strategic investment from Bright Data.
- Tangos deploys AI agents to investigate financial crime and generate regulator-ready case files.
- Founded by serial entrepreneur Eyal Azoulay, the company brings together former US Treasury sanctions officials, intelligence leaders and AI experts to tackle one of banking’s biggest operational bottlenecks.
Banks have spent the past decade teaching software to spot suspicious transactions. Tangos believes the harder problem is what happens after the alert fires, and it wants an AI agent to handle that too.
The Tel Aviv-based startup, known for the autonomous investigation layer between transaction monitoring systems and a defensible, examiner-ready outcome, has raised $20 million in seed funding led by Red Dot Capital Partners, with Leaders VC, SignalFire, Venture Israel, Clarim, Clutch Capital, and Selah Ventures joining, plus a strategic investment from Bright Data.
“Financial crime has evolved into a network problem that increasingly exceeds the capacity of traditional investigative processes. Organisations have made tremendous progress in detecting risk, but the investigation process remains one of the largest operational bottlenecks in financial crime prevention,” says Eyal Azoulay, founder and chief executive of Tangos.
The 95% problem
Financial crime generates more than $1.5 trillion in illicit proceeds a year, and compliance teams are already drowning in alerts. Investigators still have to determine, case by case, whether a flagged transaction is genuinely dirty money, trace sanctions exposure, untangle ownership, and produce a report a regulator will accept — manual work that leaves banks able to chase down only a fraction of the cases their own systems surface.
TFN’s April 2025 coverage of RegTech startup ComplyStream called this “the 95% problem.” Tangos is chasing the same gap with a more autonomous approach.
Azoulay founded Tangos in 2025 after three prior exits, one of which was acquired by BNY Mellon. His team includes former officials from the US Treasury’s Office of Foreign Assets Control, former Israeli intelligence officers, and engineers who have led large-scale autonomous AI systems.
How the AI agents work
The startup deploys AI agents that run an investigation end-to-end: pulling evidence, testing hypotheses, mapping relationships between entities, and producing a case file with an audit trail. An agent might take a sanctions alert on a shell company, trace its ownership across jurisdictions, and hand a human investigator a finished report rather than a blank page.
That is a different approach than the one most of Tango’s well-funded competitors are making. Feedzai, which raised a $200 million Series D, and ComplyAdvantage, backed by more than $100 million in funding build tools that get better at spotting risk in the first place. Tangos is automating the investigation itself, which means trusting an AI agent’s judgment, not just its pattern-matching.
Who signs off when the AI is wrong
A false positive from a monitoring tool costs an analyst a few minutes. A flawed AI-generated case file that reaches a regulator is a different category of failure. Industry analysis of autonomous agents notes that autonomy changes how an action occurs, not who is held accountable for it.
Neither Tangos nor its bank clients have publicly said how that accountability line is drawn if an investigation turns out to be wrong.
“Tangos has developed a category-defining platform that enables organisations to investigate complex cases more efficiently while maintaining the governance, transparency and evidentiary standards required by regulators,” says Yaniv Stern, co-founder and managing partner at Red Dot Capital Partners, whose past exits include Global-e, which went public on Nasdaq in 2021.
The financial crime and compliance technology market is expected to reach $26.8 billion by the end of 2025, with the narrower AML software segment alone projected to grow from $2.58 billion in 2025 to $6.78 billion by 2034, according to Fortune Business Insights. Tangos says the funding will expand its research team and deployments, though it has not disclosed headcount or hiring targets.
Whether banks let AI agents write the evidence file, rather than just point to where to look, will decide if this becomes the next layer of RegTech automation or stays a faster way to produce a first draft a human still has to rewrite.
