Catapult handles your most ambitious knowledge work
By the time you give it your first task, Catapult has already spent hours building an understanding of you and everything you've worked on, across your emails, messages, meetings, documents, and tools.
So it knows how you write. How you make decisions. What you’re building and why. The history behind every project. What needs to happen next. It feels like it has been with you from day one.
That understanding deepens as you work, compounding from one task to the next. No explaining yourself. No markdown files to maintain. No skills to write.
Fewer first drafts. More finished work.
In our internal tests, Catapult outperforms the highest-tier models in Claude Cowork and ChatGPT Work on most knowledge work tasks. Examples include:
- Answering complex questions across your work by tracking down the data, decisions and history scattered across meetings, messages, documents and tools.
- Drafting emails in your voice after reading the whole thread, understanding the relationship and checking every promise still outstanding.
- Creating social posts from work you had forgotten about, finding the strongest product progress, customer proof, conversations and lessons before writing them in your voice.
Show moreShow less
- Turning meetings into next steps by working through what was decided, updating the plans and drafting every follow-up people promised.
- Building board packs and investor updates by pulling together the latest financials, company metrics, progress, risks and commitments from previous meetings.
- Spotting which leads to contact after every release, matching what shipped to what each person previously asked for and drafting the messages.
- Drafting PRDs after reading the codebase, listening to user calls and tracing the decisions that shaped the product.
- Preparing you for customer meetings by remembering the full relationship, what each person cares about, what you promised and what remains unresolved.
- Drafting proposals and statements of work by turning customer conversations, pricing, product capabilities and implementation requirements into a finished document.
- Completing RFPs, RFIs and due diligence by finding the latest answers from across the company and filling in every spreadsheet and document.
- Answering questions about company performance by finding the right numbers, checking where they came from and explaining what changed.
- Creating tickets for recurring problems by pulling together the user reports, internal conversations and evidence engineers need to ship the fix.
- Updating the sales pipeline from what actually happened across meetings, emails, proposals, product usage and promised next steps.
- Turning recurring customer feedback into product priorities by finding the patterns across calls, support threads, usage data and internal conversations.
- Finding commitments that are about to be dropped, working out who is waiting on what and drafting the messages needed to close each loop.
- Planning renewals and expansions by bringing together product usage, customer requests, support history and recently shipped features.
- Working out why a project is late by retracing the original plan, changed requirements, dependencies, blockers and decisions behind it.
- Bringing new joiners up to speed on the product, its history, why decisions were made, who owns what and how work actually gets done.
- Writing stand-ups, performance reviews and progress updates from the work people actually did, including contributions nobody remembered to record.
- Completing security questionnaires by finding current answers across your architecture, policies, controls, vendor records and previous responses.
- Evaluating candidates by reading the job spec, CV, application, interview history and how you assessed similar candidates.
- Updating financial models with actuals and new assumptions, preserving the formulas and showing you exactly what changed.
- And many more.
We’re a small team of researchers, engineers and repeat founders from Meta, Amazon Research, G-Research, Oxford, Cambridge and Imperial. We’re building Catapult for ourselves and for people and teams who want to do their life’s best work, everyday.
Currently in private alpha.