Rogo

An AI-native finance workspace built for investment professionals — chat-first, agent-driven, and grounded in a firm's own documents plus licensed market data. Used across IB, PE, hedge funds, and increasingly growth equity / VC diligence workflows.

Chat-first workspace where users issue plain-English prompts (e.g., "build a multi-slide overview of deals in a sector since 2020") and receive finished outputs with source citations. Felix, Rogo's agent for finance, generates full deliverables — decks, financial models, memos, dashboards — from a single prompt and runs multi-step workflows asynchronously (including via email). Spreadsheet agent reads, explains, audits, and edits complex multi-tab valuation models; native Microsoft Excel add-in puts Felix inside the workbook. Pre-built "Quick Actions" for common tasks (company profiles, earnings comparisons, meeting-prep memos, deck proofreading); a slides annotator; and a "Library" of generated artifacts. "Memory" retains user role, conventions, and formatting preferences across sessions. Outputs are sourced, cited, and auditable, with role-based customization.

Agentic architecture: Felix orchestrates leading third-party models from OpenAI, Anthropic, and Google together with Rogo's own purpose-built/fine-tuned financial-reasoning models. (Earlier reporting described GPT-4o for chat/analysis and o1-family models for structuring and advanced reasoning; Rogo has since described a multi-provider orchestration approach.) Grounded in customer-permissioned data — internal files, CRM records, and licensed providers — with every output sourced and cited. Vendor-reported accuracy benchmarks (e.g., FinanceBench) are company claims, not independently audited.

Opaque, enterprise-oriented; multi-week implementations (security review, data-source entitlements, integration build). No public per-seat pricing — press/estimate figures only; confirm with sales.

Microsoft Office (Excel add-in; PowerPoint/Word outputs); SharePoint; CRMs (Salesforce, Microsoft Dynamics); data providers (PitchBook, S&P Capital IQ, FactSet, Preqin, LSEG, Quartr); custom MCP support; Anthropic Claude Marketplace distribution. Forward-deployed bankers/engineers white-glove deployment model.

SafeBase trust center with a strong compliance posture: SOC 2 Type II, ISO/IEC 27001, ISO/IEC 42001:2023 (AI management systems), GDPR, stated EU AI Act compliance, and a VPAT. Zero-trust, least-privilege access, end-to-end encryption (at rest and in transit), data siloing/single-tenant isolation, audit logging, and third-party penetration testing. Rogo states it never uses customer/private data to train or update its models. Every output is sourced, cited, and auditable, supporting human-oversight requirements.

AI-native finance startup; positioned among the leading vertical AI copilots for investment professionals alongside AlphaSense and Hebbia. Valuation and per-seat pricing figures circulating in press are estimates rather than officially confirmed.

Strongest fit for diligence-heavy workflows where outputs must be cited and auditable — IC memos, sector overviews, comparable analyses, earnings deep-dives, and Excel-based valuation work. The Excel add-in and spreadsheet agent are differentiated for modeling. MCP + Claude Marketplace make it composable with broader AI-native workflows. For VC/GE, fit is strongest at growth stages and at firms with meaningful internal document corpora and licensed-data subscriptions.

Opaque, enterprise-oriented pricing and multi-week implementations are likely heavy for small VC funds. Forward-deployed services model is differentiated but closer to professional services, which may affect deployment timelines. Vertical fit is narrower outside core IB/PE workflows; horizontal copilots may be cheaper per seat for general use. Highly competitive space (AlphaSense, Hebbia, FactSet Mercury, S&P/Kensho, Bloomberg); reliance on third-party frontier models is a structural dependency.