Andera, a San Francisco-based AI-native platform designed to automate internal audit workflows, has secured $37m in a Series A funding round led by Lightspeed Venture Partners.
The capital will be used to scale Andera’s platform across Fortune 100 enterprise customers, grow its engineering and customer success teams, and expand its automated controls database.
Andera’s platform automates the full end-to-end testing of controls spanning SOX, operational, and compliance categories, covering everything from user access provisioning through to complex bad debt reserve and management review controls. The system processes hundreds of millions of tokens of financial evidence per control, drawing on data from Excel workbooks, PDFs, system screenshots, and journal entries, then applies AI to surface the relevant documentation and deliver an audit judgement. The company’s longer-term ambition is to support the documentation, assessment, and testing of any type of control, regardless of industry, sector, or underlying system.
Lightspeed spent several months developing its thesis around audit automation before committing to the investment, speaking with finance and audit leaders across Fortune 500 companies in technology, financial services, healthcare, and consumer sectors, as well as Big Four partners and former PCAOB officials. The firm concluded that internal audit had not been meaningfully reimagined for two decades, and that the convergence of large language model reasoning, long context windows, and agentic tool use had created conditions for genuine automation where earlier rules-based approaches had failed. The average public company spends $3m annually on audit fees, while Fortune 100 businesses can pay upwards of $20m a year.
Andera was founded by Aryo Patel and Tinah Hong, who met in Chicago as children and studied together at MIT. Patel brings experience from Azure infrastructure scaling at Microsoft and the trading desk at Jane Street. Hong, who serves as CTO, previously built machine learning systems at Stripe and worked as a software engineer at Microsoft. The company’s technical architecture includes a proprietary agentic retrieval system designed to surface the right evidence across billions of tokens, probabilistic beam search that directs additional compute towards areas of uncertainty in the data graph, and custom agents that manipulate financial workpapers with the precision required by large enterprise clients. The team has also brought in seasoned industry practitioners, including Carina Averilla, a former Deloitte auditor who served as head of US accounting at Revolut, and Jared Lauber, who leads partnerships and has spent 30 years as a risk and audit leader at organisations including EY, Instacart, McKesson, and Williams-Sonoma.
Lightspeed noted that CFOs at publicly traded companies are under mounting pressure to achieve efficiency gains of between 200% and 300% from back-office functions. The firm also argued that the Big Four face a structural tension: their business model, which depends on billing hours, makes it difficult to develop the best AI products without reducing their own headcount. Lightspeed’s view is that this opens a window for emerging players to build at the cutting edge and position themselves as natural partners for the large firms.
Copyright © 2026 RegTech Analyst
Copyright © 2018 RegTech Analyst
