Beyond the pilot: Dyna.Ai raises an eight-figure Series A to put agentic AI to work in financial services
The pilot era in finance is losing steam. Dyna.Ai, a Singapore-headquartered AI-as-a-Service company, has closed an eight-figure Series A led by Lion X Ventures (advised by OCBC Bank's Mezzanine Capital Unit), with participation from ADATA, a Korean financial institution, and several industry veterans. The capital will accelerate deployment of Dyna.Ai's agentic AI platform already active across banks and financial institutions in Asia, the Americas, and the Middle East. The pitch is simple: execution, not experimentation.
The pilot problem in finance
Banks spend months on proofs-of-concept, rack up dashboards, and then stall at compliance gates. The sticking points are predictable: audit trails, approvals, data residency, and integration with core systems. Agentic AI adds another layer-these systems don't just suggest; they act within set parameters. Without governance baked into the product, nothing scales.
Execution over experimentation
Founded in 2024, Dyna.Ai kept its scope narrow by design. It builds domain-specific agent builders, task-ready agents, and fully operational applications that slot into existing workflows. The "Results-as-a-Service" model is built for regulated environments where outcomes, controls, and timelines matter. "While much of the industry was focused on how broadly AI could be applied, we doubled down early on a specific, pressing problem and built it with outcomes in mind," said Dyna.Ai chairman and co-founder Tomas Skoumal.
Why investors are leaning in now
Across the region, the question has shifted from "should we use AI?" to "how do we make it stick?" "Enterprise AI is entering a phase where execution and measurable outcomes matter more than experimentation. Dyna.Ai differentiates itself through strong domain expertise, operational discipline, and the ability to deploy agentic AI within complex, regulated enterprise environments," said Irene Guo, CEO of Lion X Ventures.
The bar is high. In banking and insurance, agents must trigger workflows, update records, and handle documentation with full accountability. That calls for governance architecture built into the product-aligned to principles like the MAS FEAT principles-not bolted on after a pilot. As Cynthia Siantar, Dyna.Ai's Head of Investor Relations and General Manager for Singapore and Hong Kong, put it: "The focus has moved past pilots and experimentation to how AI can be deployed in day-to-day operations and deliver real outcomes."
Where agentic AI already fits in a bank
- Customer operations: KYC refresh, case summarization, email triage, and document chasing with audit logs.
- Credit operations: Document intake, data extraction, exception handling, and rule-based approvals with human checkpoints.
- Compliance: Name screening triage, evidence collection, policy-aware case notes, and immutable activity trails.
- Finance and back office: Reconciliations, variance explanations, report drafting, and ledger updates with maker-checker controls.
- Collections and retention: Next-best action under guardrails, compliant communications, and CRM updates.
The common thread: agents act inside defined workflows, respect entitlements, and leave a clear audit trail.
What to verify before you scale
- Controls: Guardrails mapped to policy and regulation, approval workflows, and segregation of duties.
- Auditability: End-to-end logs, case replay, and tamper-evident records tied to user IDs and data sources.
- Data: PII handling, redaction, residency, and encryption standards that meet internal policy and regulator expectations.
- Model risk: Documented testing, drift monitoring, fallback paths, and alignment with your model risk governance.
- Human-in-the-loop: Clear thresholds for escalation and overrides, with SLAs.
- Economics: Cost-to-serve, cycle-time reduction, quality lift, and time-to-production measured in weeks, not quarters.
A market that's ready
Macroeconomic pressure is pushing institutions to ship outcomes, not slide decks. The investor mix here-OCBC-advised capital, a Korean financial institution, and a Taiwan-listed technology company-signals cross-border demand across both buyers and infrastructure partners. Financial services remains a high-value target for agent-based automation precisely because controls are non-negotiable-and that's where specialists earn their keep. The pilot shelf life is shrinking; production-grade governance is the new baseline.
For finance leaders: a simple plan
- Pick one high-friction workflow tied to a revenue or risk KPI (e.g., KYC refresh time, exception backlog, claim cycle time).
- Codify guardrails and approvals first, then plug in the agent. Governance before scale.
- Commit to a 60-90 day go-live with a small set of measures: cycle time, error rate, and unit cost.
- Treat it as an operating service with SLAs, not an endless pilot.
If you want structured playbooks and case studies, see AI for Finance.
Your membership also unlocks:
