Intapp makes its Celeste AI coworker generally available
Most professional firms already have AI on every desk. Drafting is faster. Modeling is faster. Growth often is not. Intapp is selling that gap as a product category. On July 15, 2026, the company made Celeste generally available — an “expert AI coworker” meant to run the business of the firm, not just polish the work product.
Celeste was previewed in February. GA means the pitch moves from stage demo to procurement and production workflows across law, investment banking, private capital, accounting, consulting, and real assets — the regulated lanes Intapp already owns.
What GA unlocks vs earlier pilots
Per Intapp’s investor announcement, Celeste is grounded in each firm’s proprietary knowledge and governed by the firm’s own methods and compliance rules. It is built to run on its own rather than wait for a prompt on every step. The company ships a library of playbooks — multi-step workflows encoding firm best practices — spanning origination, business development, intake, conflicts, pricing, fundraising, lateral hiring, and more, plus a no-code builder for custom playbooks.
Native integration into systems of record matters. Celeste ingests firm documents and connects to preferred data partners through open connectors. The differentiation claim is not a better base model. Every firm can buy similar models. Intapp argues the edge is deals, clients, relationships, and prior judgment that generic assistants do not hold under firm control.
Early-adopter names in the release include Hg and BakerHostetler. GA still has to survive conflicts teams, infosec, and partners burned by confident wrong answers.
Conflicts and clearance workflows
Conflicts is where Celeste either becomes infrastructure or another dashboard. Intapp describes end-to-end lateral hire flows from candidate research and conflicts pre-screening to clearance preparation and onboarding. In private capital, Celeste turns a confidential information memorandum into a structured investment screen against the firm’s mandate and deal history. In legal, it surfaces relevant matters, partner experience, and prior pitch materials for urgent client requests in minutes.
BakerHostetler CIO Katherine Lowry said Celeste can positively impact intake and conflicts amid heavy lateral growth. Hg’s Shirin Veeran stressed screening against Hg’s actual mandate and knowledge built over two decades — “ours, not something generic.” Those quotes track Intapp’s thesis: Firm AI beats horizontal AI when the work is relationship- and obligation-shaped.
Thad Jampol, Intapp co-founder and CPO, went further: when a conflicts team can clear 10 times the matters and a fund can screen 10 times the deals, that is a new business model, not saved time.
Governance hooks for regulated firms
Intapp’s brand is governed AI. Celeste is supposed to inherit ethical walls, need-to-know rules, and audits the same way people do. Restricted information stays restricted regardless of the prompt, and firms can show clients and regulators a full record of every action. That is the procurement paragraph. Without it, no Am Law or bulge-bracket buyer stays in the room.
Intapp calls the category Firm AI: not horizontal assistants for everyday tasks, not practice AI for work product, but systems that run the business behind both.
Who should evaluate Celeste first
Start with firms that already run Intapp for intake, conflicts, or relationship intelligence and feel the bottleneck in clearance, deal screening, or lateral onboarding. Private capital teams drowning in CIMs, law firms with lateral surges, and banks fighting relationship amnesia are the natural pilots.
Ask for playbook coverage in your practice, audit logs your risk team will accept, and a conflicts test on real restricted matters.
Analysis: Intapp is not trying to win chat. It is trying to own the operating system of regulated firm growth — conflicts, intake, origination, hiring — under rules horizontal tools were never built to respect.
