Pinegap raises $8m to deploy AI agents across buy-side funds

Pinegap, a New York-based equity research automation platform founded in 2024, has closed an $8 million Series A led by Stellaris Venture Partners, with existing backers Inventus, Silicon Valley Quad, and DeVC also participating. The funding will be used to expand its go-to-market function, grow its engineering team, and recruit a bench of former buy-side analysts to deepen its domain capabilities.

The company builds custom AI agents for institutional buy-side investors, hedge funds, long-only mutual funds, and registered investment advisors, tuning each deployment to a fund's investment style, proprietary data, and preferred output formats. Unlike general-purpose AI assistants, Pinegap's agents are push-based: outputs arrive in an analyst's inbox on a set schedule or when triggered by market events, without requiring a user query. The company says it has deployed more than 1,000 agents across more than 100 institutional clients and is now generating upwards of 50,000 research reports per month.

Automating the analyst's day

The founding logic is straightforward. Buy-side analysts typically maintain active coverage of 20 to 40 companies while monitoring several hundred more on a watchlist, and the workflows that structure their day, earnings previews, company primers, thesis-tracking updates, are largely executed by hand. Co-founder and CEO Deepak Sharma, an Indian Institute of Technology alumnus, describes the platform's ambition plainly: "Every workflow we automate, whether it's an earnings preview, a company primer, or a thesis tracker, is time an analyst gets back to focus on what only they can do: judgment, conviction, and decisions. Pinegap isn't a chatbot or a search tool. It's a platform built around how a fund actually operates, tuned to their data, their format, and their investment style."

The company was co-founded by Ankit Varmani, a 15-year institutional finance veteran with JPMorgan experience who was preparing to launch his own fund when the partnership began. Stellaris partner Alok Goyal framed the investment case around a gap between the intellectual demands placed on buy-side professionals and the tools available to them, describing Pinegap's team as combining "the domain depth of a career analyst with the technical velocity of a product builder."

The broader convergence at stake

The raise sits within a rapidly expanding contest to capture the AI infrastructure layer of institutional asset management. Pinegap is not alone in the space: Bloomberg is embedding generative AI into its Terminal workflows, and a cluster of enterprise AI vendors, including Visible Alpha and Sentieo, have built research-aggregation tooling for institutional teams, though none yet claims the same degree of per-fund agent customisation that Pinegap is pitching. The distinction between a queried AI tool and an autonomous, scheduled agent that runs independently is a meaningful one for compliance-conscious funds: the audit trail and explainability of push-based outputs will become a regulatory focal point as adoption broadens.

The macro context matters for cross-sector strategists. Asset management is a data-intensive industry under structural cost pressure: fee compression from passive investing and rising data-vendor costs are squeezing the economics of fundamental research. AI-driven automation offers a partial answer, but it also accelerates a longer-term question about headcount in the analyst cohort. If a single platform can generate 50,000 research reports monthly across 100 clients, the downstream implications for financial services talent pipelines, and, further out, for the graduate recruitment strategies of the major investment banks that feed those pipelines, are non-trivial.

For capital allocators watching the enterprise AI buildout more broadly, the Pinegap round is a small but illustrative data point: early-stage vertical AI platforms targeting high-value professional workflows are still attracting Series A capital, even as the market debates whether foundation-model incumbents will eventually commoditise the same functionality from above.