Audit technology startup Denki has raised $4.1 million in funding to modernise the way financial audits are conducted. The round was led by Base10 Partners and Shine Capital, with participation from Y Combinator, 20VC and other investors.
Talking to TFN, Felipe Jin Li, the co-founder at Denki, stated: “We are not disclosing valuation at this time.”
Why is the audit industry ready for change?
Financial audits form a crucial part of the global economy’s trust framework. They ensure financial statements are accurate, confirm that internal controls are functioning properly, and provide investors with confidence in the company’s disclosures.
Despite this central role, much of the work still relies on methods that resemble processes from the early 2000s. Auditors often spend long hours gathering evidence, organising documentation, and manually verifying controls across multiple systems.
At the same time, pressure on the industry is intensifying. Regulators are demanding stronger oversight and greater accuracy, while firms face growing complexity in financial reporting. In the United States, the Public Company Accounting Oversight Board issued $35.7 million in penalties in 2024, reflecting heightened scrutiny of audit quality.
Meanwhile, major accounting firms are reshaping their workforce strategies. PwC has announced plans to cut entry-level hiring by roughly one-third by 2028 as routine tasks shift toward automation and offshore operations.
Together, these trends have created a moment when new tools are no longer just helpful but necessary.
Built by engineers who understand audit work
Denki was founded by brothers Felipe Jin Li and David Jin Li, who combined technical expertise with experience in finance and research.
Before launching the company, Felipe worked as a PhD researcher in Explainable AI at University College London after earlier experience at McKinsey & Company. David studied computer science at Imperial College London and previously built financial data pipelines at MacroHive, used by leading hedge funds.
Speaking about the motivation for starting this company, he said, “Denki was founded to modernise financial audits and address the ‘leverage problem’ in audit teams.”
Founders observed that audit teams spend 80% of their time on manual evidence collection and only 20% on the exercise of skilled judgment. They recognised this as a market need, rather than a purely personal experience.
Their shared belief was that the best audit software should be built by engineers who genuinely understand how audit work happens in practice.
Denki also took part in the Fall 2025 batch of Y Combinator, helping refine the platform while connecting with early customers and investors.
Turning audits into structured, collaborative workflows
Denki’s platform aims to transform how auditors interact with data and documentation. Instead of relying on fragmented spreadsheets and static files, the system treats audit procedures more like structured software processes. The company is building a collaborative platform designed specifically for auditors working with public companies.
The platform allows audit teams to review financial evidence more quickly, organise documentation more intelligently and test controls with greater precision. By introducing automation and collaborative tools, Denki hopes to make audits reproducible, testable and easier for teams to manage across large organisations.
This approach reduces time spent on repetitive administrative tasks while allowing auditors to focus on judgment-heavy analysis and risk assessment.
Denki is designed primarily for audit teams working with pre-IPO and publicly listed companies, where compliance demands are highest, and evidence requirements are extensive.
How is Denki different from the rest?
The co-founder told us, “Denki is a collaborative AI platform for auditors. Key differentiators are it automates manual, evidence-heavy processes; makes audits reproducible, testable, and collaborative; enables full-population testing instead of sampling; integrates directly with financial systems for automated evidence collection; and focuses on freeing auditors to do skilled judgment rather than repetitive tasks.”
Looking ahead
Detailing the plans for the next three to five years, he stated the following:
- Hire engineering talent with audit and finance backgrounds.
- Continue building the platform to help auditors collect evidence faster, document smarter, and test controls with precision.
- Position Denki as the modern infrastructure standard for financial audits, especially as AI adoption grows in the industry.
“The best audit teams don’t have a talent problem—they have a leverage problem,” said Felipe Jin Li, co-founder and CEO of Denki. “They spend 80% of their time pulling evidence and matching documents, which leaves 20% for the work that requires actual expertise: interpreting exceptions, assessing risk, advising leadership. At Denki, we believe that the ratio should be flipped.”
His brother and co-founder, David Jin Li, added, “When we sat with audit teams, we saw two very different types of work. Deciding what a control failure means, what to escalate, how to advise management—that’s skilled judgment. Pulling files out of SAP and checking that timestamps match? That’s a computer problem. The manual work exists because no one built the infrastructure to automate it.”
“Sampling exists in audit because it’s historically been too expensive to analyse everything. That constraint no longer applies,” said Randy Lund, Chair of Columbia Banking’s audit committee and former KPMG audit partner. “Denki is rebuilding audit workflows around full-population testing, automated evidence collection, and system-level integrations.”
“Felipe and David are building Denki at exactly the right moment, when the biggest players in this industry are publicly acknowledging that the old model is breaking,” said Caroline Broder, Partner at Base10 Partners. “Denki’s platform will help audit teams collect evidence faster, document procedures automatically, and test controls with greater precision across the systems where financial data actually lives.”
