Working now
Reviewing the Q2 policy draft
Checking the approval rule, preparing an email summary, and logging decisions.
- Calendar
- Drive
- Slack
- GitHub
- CRM
Clerwell – hire your first AI worker
Hire your first AI worker
Give them a name. A role. An email address. Your guidelines. Then give them work.
Clerwell creates AI workers who live inside your digital workspace. They join meetings, read files, respond to emails, update systems, remember instructions, ask for approval, and carry work forward. Not models you call – workers you onboard.
Activity feed
- 09:42Joined policy review
Calendar invite accepted with agenda attached.
- 09:51Drafted email
Prepared summary for Finance and Legal.
- 09:58Asked approval
Routing exception to Priya before sending.
- 10:01Logged action
Decision record saved to the workspace.
What is Clerwell?
A working identity for the role you need
Give Clerwell a job description and we give that role a working identity – a name, an email, memory, tools, permissions, accountability. The worker learns your environment, connects to your systems, and gets on with the work.
Identity matters because work gets easier when you know who owns what. Every worker has a name, title, email, calendar presence, Slack or Teams profile, role-specific memory, and a defined set of tools. You send a document to Arjun, ask Meera to check a script, invite Neha to a review. No prompts, no endpoints, no workflow builders. Just people.
Neha Menon
WorkingDocument Review Worker
Arjun Mehta
WaitingLegal Ops Worker
Meera Kapoor
ReviewingDeveloper Assistant
Onboard them like employees
- 01Name and role
Email, title, calendar presence, and owner.
- 02Guidelines
Policies, SOPs, examples, tone, and preferences.
- 03Connected tools
Email, calendar, Slack, Teams, Drive, GitHub, Jira.
- 04Boundaries
Permissions, approval rules, and escalation paths.
- 05First assignment
Review the policy draft and prepare follow-ups.
They work inside your workspace
Workspace map
EmailCalendarDriveSlackGitHubCRM
Slack / Teams
Approval queue
- Policy noteNeeds approval
- Email draftReady
- Decision logSaved
They learn how you work
Guidelines, policies, SOPs, tone of voice, examples, approval rules, and preferred documents stay with the worker.
The difference between AI platforms and AI workers
AI platform
- Ask
- Reply
- Copy
- Decide
- Send
- Track
Clerwell worker
- Reads
- Checks context
- Asks approval
- Sends
- Logs
- Follows up
One produces answers. The other carries responsibility through your systems.
Any role, defined by you
You do not fit into rigid categories. Neither should your workers. Just give Clerwell a job description.
Inbox manager
email@yourcompany.com
Sorts daily emails, flags urgent requests, and drafts routine replies based on past correspondence.
OnlineContent strategist
content@yourcompany.com
Audits rough drafts, finds gaps, suggests structural improvements, and tracks publishing progress.
PlanningDeveloper assistant
dev@yourcompany.com
Reviews commits against your personal style guide and flags optimization issues.
ReviewingFinance assistant
finance@yourcompany.com
Matches invoices with past receipts and flags unusual expenses.
WaitingDescribe the role. Provide the guidelines. Clerwell gives you the worker.
Start with one worker
Pick one role where work is repeated, delayed, or fragmented. Give them a name, context, tools, boundaries, and work. Then hire the next one.
- 01Pick one role
Choose repeated, delayed, or fragmented work.
- 02Name the worker
Give ownership a clear identity.
- 03Provide context
Guidelines, tools, documents, and examples.
- 04Set boundaries
Permissions and approval rules.
- 05Assign work
Let the worker carry it forward.
- 06Hire the next one
Repeat with the next role.
Trained on your way of working
Everyone has their own language, approval rules, writing style, and repeated decisions. Clerwell can train small language models on your approved knowledge. Routine work gets handled by your SLM. Complex work routes to your chosen providers. Sensitive work stays entirely local.
The worker does not run to an external model for every small call. It learns your ordinary rhythm.
Self-deployed by default
Your AI workforce runs under your control. Every plan is self-deployed. Every plan supports unlimited workers. Bring your own keys. Choose your own model provider.
If you want help, we offer deployment support, managed deployment, custom SLM training, and advanced controls.
Hire the worker you need next
Start with one. Name them. Give them a role, your rules, tools, boundaries, and work. Then let them show up tomorrow with the context they learned today.