Sylvan Labs sells $10.94M offering for autonomous revenue teams

Why it matters

Sylvan's full offering gives two enterprise-software veterans the capital to test whether AI agents can own post-sale revenue work, where bad data carries immediate customer risk.

Charles Li (@charlesli09) and Shangyan Li (@ShangyanL) founded Sylvan Labs, which has sold a $10,940,500 equity offering for its year-old attempt to give AI agents responsibility for growing revenue from existing customers.

The Form D filed with the SEC on September 2nd shows that Sylvan sold the full offering to 34 investors, leaving no securities unsold. The first sale occurred on August 3rd, meaning the financing had been underway for about a month before Sylvan disclosed it.

The filing does not name a lead investor, disclose a valuation or classify the financing as a seed or Series A. A current Sylvan hiring page says Sylvan has raised more than $10 million from investors including TQ Ventures and HubSpot Ventures. That page does not establish whether either investor participated in this offering.

Sylvan has not published revenue, customer-count or retention figures, so the filing does not provide enough information to assess the company's operating traction.

Two founders coming from opposite sides of enterprise software

Charles Li brings the investor's view of the enterprise software market. Before starting Sylvan in 2025, he founded and ran college-admissions consultancy V2 Admissions, worked in software private equity at Vista Equity Partners and spent time at Bessemer Venture Partners. Sylvan puts him on the other side of the financing table, selling a product into the same class of recurring-revenue businesses that private equity firms have spent years measuring and optimizing.

Shangyan Li brings the operating history. The Harvard computer science graduate previously worked as a software engineer at Ripple and co-founded Butter Technologies with Winston Chi in 2020. Butter built software for food distributors, using AI to convert orders arriving through voicemails, texts and other unstructured channels into data that operators could review and process.

GrubMarket acquired Butter in 2024, folding its eight-person team and software into GrubMarket's food-distribution platform. Butter had raised about $12.3 million, including a $9 million Series A at an approximately $39 million post-money valuation. The acquisition price was not disclosed.

That experience supplies the practical logic behind Sylvan. Butter had to turn fragmented, industry-specific information into actions inside a customer's existing workflow. Sylvan is applying a similar model to customer success and account management, where useful signals are split among warehouses, product-usage logs, CRM records, support conversations and billing systems.

Sylvan wants the work after the sale

Sylvan describes its product as a foundation for training, deploying and optimizing revenue agents. Its focus is the installed customer base rather than cold outbound: detecting churn risk, identifying expansion opportunities and helping revenue teams pursue upsells, cross-sells and growth following customer acquisitions.

Shangyan Li outlined an earlier version of that data layer when Sylvan introduced the Signal Library on August 27th, 2025. Sylvan said the system unified scattered customer inputs into structured signals for revenue and customer-success teams.

Sylvan's current engineering materials offer a more technical description. Sylvan is building an ontology and semantic layer that maps each customer's warehouse into a common representation, with a signal pipeline and agent platform operating above it. A deployment engineer is expected to inspect customer warehouses, translate bespoke data models and validate the queries that agents use to make decisions.

That detail matters because revenue agents have little room for plausible-sounding mistakes. A model recommending the wrong expansion play can annoy a customer. An agent acting on faulty churn data can create a problem where none existed. Sylvan's public materials say its agents act across customer accounts, while leaving the approval controls and degree of production autonomy unspecified.

Sylvan says its customers range from fast-growing startups to Fortune 500 businesses and names ServiceNow in its hiring copy. Sylvan has not published a case study or measurable outcome supporting that customer claim.

The hiring page also shows where part of the new capital is likely headed. Sylvan is recruiting an engineering leader at a base salary of $200,000 to $300,000, with responsibility for the agent platform, machine-learning strategy and additional hiring. The SEC filing lists Sylvan's principal address in San Francisco, while the job page describes Sylvan as fully in-office in New York City.

Revenue agents are becoming a funded category

Sylvan's financing lands in a crowded market. Attention announced a $30 million Series B in June for agents that handle sales follow-ups, CRM updates and other actions after calls. Alta followed with a $25 million Series A in July for a broader network of agents covering prospecting, qualification, account management and cross-selling.

Sylvan's wedge is narrower and potentially valuable: finding more revenue in customers a business has already paid to acquire. Its technical bet is that a shared model of proprietary customer data can make agents useful across the entire account base, rather than turning each workflow into another isolated automation.

The market still has to prove that those agents improve results. Gartner said in July that fewer than 40% of sellers will report productivity gains from AI agents by 2028. Gartner warned that fragmented systems can cause agents to scale the fragmentation instead of fixing it.

Charles Li and Shangyan Li have raised ahead of that reckoning. The $10.94 million gives them room to hire the engineers and deployment staff required to reconcile enterprise data one customer at a time. The harder job is producing evidence that Sylvan's agents can find and protect enough revenue to justify another system inside an already crowded sales stack.