Finpilot raises $10 million Series A to bring AI agents to endowments

Finpilot, a Seattle start-up that builds artificial intelligence software for institutional investors, announced on Wednesday, 4 November 2026 that it has raised $10 million in a Series A round. The round was led by Lobby Capital, and Madrona, which led the company's seed round in 2024, invested again. Founder and chief executive Lakshay Chauhan said the money will be used to expand the company's AI agents and to hire across engineering, product, client success, sales and marketing.

Institutional investors such as endowments and pension funds look after very large pools of money, but much of their daily work is still done by hand. Their investment teams receive reports, statements and legal documents from dozens of outside fund managers, often in different formats, and staff spend long hours pulling numbers into spreadsheets before anyone can analyse them. Finpilot is betting that this paperwork is a good fit for AI agents, which are software programs that can carry out a series of steps on their own, such as reading a document, finding a figure and adding it to a report, rather than only answering a single question.

The company describes its platform in three layers. The first is a single, accurate record of everything the organisation owns and every document it holds. The second is a model of how that data connects. The third is the set of agents that do the work on top of it. The idea is that an agent is only as reliable as the data it reads, so the record has to be right before the automation is useful. That matters in this field because a wrong number in a report to a pension board or a hospital's investment committee is not a small error. These are the company's own descriptions of its product; independent tests of how accurately the agents work have not been published.

The customer list is also worth noting. Several named users are hospital systems and health organisations, which often run large investment pools to support their operations. A state retirement system and an outsourced investment office are also on the list. For a company at the Series A stage, a set of named institutional clients helps explain why an existing investor chose to put in more money.

For readers in India, the story shows where AI money is moving in finance. A great deal of attention goes to chatbots for retail customers, but this round is aimed at the back offices of large investors. India has a growing base of family offices, insurance funds and pension managers, and the same problem of scattered documents and manual reporting exists there. Software of this kind is one of the areas where Indian engineers and founders already compete for global customers.

Peepals timeline: Chauhan has worked in machine learning since 2016 and, before starting Finpilot, built machine learning and quantitative research work at the investment firm Euclidean, according to Madrona. Madrona led Finpilot's seed round in 2024. On 4 November 2026 the company announced its $10 million Series A, led by Lobby Capital with Madrona joining again, and said it had spent three years working on the precision and accuracy of its system.

Short-term: Finpilot plans to hire across engineering, product, client success, sales and marketing, and to add more agent features to its platform.

Long-term: if AI agents prove accurate enough for investment reporting and due diligence, institutional investors could shift staff time from collecting data to analysing it. The test will be whether clients trust the software's numbers in reports that go to boards and regulators.

Who is affected: investment staff at endowments, pension funds, foundations, hospital systems and family offices; firms that sell older reporting software to these investors; and founders building AI tools for finance, including those in India.

Questions and Answers