Mantle Helps Founders Understand Their Cap Tables and Term Sheets

01 /The Concept

Most founders are not lawyers or finance specialists – and the documents that govern their ownership stakes are written by people who are. Term sheets, convertible notes, option pools, dilution waterfalls: these are instruments with enormous downstream consequences that founders routinely sign without fully understanding what they have agreed to.

Mantle has built a platform that sits in that gap. The core is an AI assistant trained specifically on the documents, concepts, and decisions involved in startup cap table management and fundraising – a narrow enough domain that the hallucination problem plaguing general-purpose AI assistants becomes tractable.

At its simplest, the assistant works like a knowledgeable chat interface. A founder can ask "how will my equity change if I sign this convertible note?" and get a specific answer – either from a general framing or, if the founder uploads the actual term sheet, from the concrete terms in that document. The platform can model what-if scenarios: what happens to equity at various valuation caps, how a new round affects existing investors, what triggers an anti-dilution clause.

Beyond Q&A, Mantle automates the procedural side of equity management. A founder can instruct the assistant to issue options to key employees on specified terms; the system prepares the document set, routes it to board members for approval, collects electronic signatures, sends executed agreements to the employees, and reports back. Tasks that previously required a law firm to run point on can now be triggered by a founder with no legal background.

The document analysis module handles the complexity of term sheet review. Upload a multi-page document with bespoke conditions and nested clauses – the assistant extracts the economic and control implications, surfaces them as plain numbers and plain-English summaries, and answers follow-up questions without requiring the founder to parse the legalese first.

The platform isn't limited to the fundraising moment. It covers option allocation, financial modeling, hiring plan scenarios, and board document management, with all materials stored in one place rather than scattered across email threads and spreadsheet folders. Recurring workflows – board resolutions, option grants, investor updates – can be templated, scheduled, and tracked through a single dashboard.

Mantle was founded last year and has now closed its first funding round: $7.68 million (CAD $10.5 million).

02 /Why It Matters

Mantle sits at the intersection of three trends that are independently generating significant investment.

The first is the rise of domain-specific AI assistants. General-purpose models like ChatGPT are useful but unreliable in high-stakes professional contexts: they hallucinate statutes, invent precedents, and confabulate terms. For a chatbot giving lifestyle advice, that's acceptable. For one advising a founder on a term sheet, it isn't. The solution is Retrieval Augmented Generation – pairing the language model with a curated, maintained knowledge base that the model cites rather than invents. Building and maintaining that knowledge base for a specific domain is itself a meaningful technical moat. Comparable startups that have attracted investment on this basis include Mindtrip (travel planning, $7M pre-launch), Curipod (lesson planning for teachers, $4.8M), Sizzle (homework help, $7.5M), and Rippey.AI ([covered previously](/review/ne-tuda-gde-kruto-a-tuda-gde-dengi), $4.8M in logistics AI).

The second trend is AI's penetration into legal workflows. Contracts are structurally complex documents where small wording choices carry large economic consequences. AI that can instantly parse a dense agreement, extract all material provisions, check them for internal consistency, and answer questions about specific clauses is genuinely useful in a way that general-purpose summarization isn't. The Contract Network ([covered here](/review/ne-tuda-gde-kruto-a-tuda-gde-dengi)) raised $8 million for a similar AI-native legal platform.

The third trend is AI as a force multiplier for lean founding teams. Mantle replaces the need for in-house legal and finance staff at the early stage. In the same spirit, Gobi has built a platform where an AI substitute for a product manager helps founders draft a product roadmap, then connects them to a marketplace of contractors for execution – and attracted $400,000 before launch.

03 /Opportunities

Startup equity management is a real problem, but a narrow market. Most startups fail early, which limits how long any given customer relationship lasts and caps the aggregate addressable revenue.

The larger opportunity is the general pattern: specialized AI assistants in domains characterized by complex documents, high-stakes decisions, and significant information asymmetry between the professional and the buyer. Legal and logistics meet all three criteria. Lesson planning meets the stakes criterion without much document complexity, but requires near-perfect accuracy because the downstream consumer is a student.

The build pattern is now well understood: identify the domain, assemble and maintain the knowledge base, connect it to a capable general model, wrap it in a workflow that replaces a manual professional process, and price it against the cost of the professional it displaces. The open question is which domains still have room for a first-mover with a strong knowledge base – before the category gets crowded. That specific constraint is the thing worth mapping before choosing where to build.