mudpie

Company profile · 3 min read

Dock gives humans and proactive AI agents one shared operating workspace

Dock coordinates role-based agents, memory, tools, workspaces and handoffs so teams can delegate recurring work without managing isolated chat windows.

Published · Updated

Dock is a workspace for teams that want multiple AI agents to behave like a coordinated team rather than a row of isolated chat windows. It is aimed at founders and operators who are willing to give agents persistent roles, tools and memory, then supervise the work through a shared record.

What it does

Dock’s current site shows agents with distinct roles working across workspaces, conversations, files and launch plans. The company says agents can pursue goals, message each other, hand off work, keep memory and continue operating in the cloud or on a local machine. The help center describes roles, tools, workspaces, email/iMessage access, shared files and human-visible activity. Dock homepage Dock help

The advantage is coordination. A founder can ask a chief-of-staff agent to keep context, let a growth agent work a list and have a content agent prepare the launch material without manually relaying every update. That is different from giving each agent a separate prompt. The tradeoff is that the system can act across more surfaces, so ownership, permissions and approval need to be explicit.

The public site is in early-access mode and does not publish standard pricing. The product examples contain illustrative activity and outcomes, but I am not treating those interface examples as verified customer results. Ask where data lives, whose credentials agents use, which actions can send messages or change files, how memory is edited, and what happens when an agent goes quiet or makes a wrong handoff.

The YC profile identifies Mike Molinet and Govind Kavaturi as former Thena co-founders, with Branch.io experience scaling to $100M-plus. That operating background matches the coordination problem, not a claim that every company should delegate its operations to Dock.

My editorial take

I would try Dock on a bounded recurring workflow—briefing, research, launch coordination or CRM hygiene—where every action can be reviewed. Keep outbound messages, payments and production changes behind approval until the team has watched the activity record. The product only earns its “team” metaphor if handoffs reduce work rather than create a new supervision job.

Quick facts

Field Sourced detail
Buyer Founders, operators and AI-native teams
Product Multiplayer workspace for proactive agents and humans
Access Waitlist and early-access product surface are published
Pricing Not published in the checked sources
Main fit question Can coordinated agents remove more context-switching than they create?

Sources checked

Dock’s YC profile, YC launch, current homepage and help center were checked on 2026-09-19.

Cohort context

Dock is listed in Summer 2026. In our 2026-09-18 directory snapshot, 119 of 232 listed companies in that cohort have YC’s primary industry label B2B (51.3%). This is a current-directory comparison, not an original intake count or a performance ranking. Nine-cohort dataset.

Public website snapshot

Observed 2026-09-19T16:18:44.460Z in raw homepage HTML. This records visible metadata and advertised links, not agent execution or product quality.

Signal Homepage observation
Product description metadata Observed
Canonical link Observed
H1 or H2 heading Observed
Typed structured data Not observed in this response
Docs/developer link Not observed in this response
Pricing link Observed
llms.txt link Not observed in this response
Markdown alternate Not observed in this response

Public observations · Collection method. Missing links here do not establish that a capability or file is absent elsewhere.

About the author

I cofound Lazyweb and publish Mudpie. This is an owner-written publication, not an independent testing organization. Research notes distinguish observations, sourced reporting and editorial judgment.

First1000 ↗ · X ↗