Does Junior use Claude?

Yes. Junior supports Anthropic Claude models, and you can select the model on a per-AI-employee basis from the dashboard.

How is Junior different from Claude Code?

Claude Code is a terminal-first developer tool — strong for developers comfortable in a CLI. Junior is a workspace coworker for whole teams — it lives in Slack or Teams, watches channels you point it at, and ships outputs without anyone opening a terminal.

Can I run Junior with my own Claude API key?

Pricing on /hire bundles model cost. Self-bring-your-own-key isn't supported today; if you need it, talk to a human via the demo form.

Is Junior faster than Claude for one-off questions?

No — for ad-hoc Q&A, opening claude.ai is faster. Junior is built for recurring, cross-tool workflows, not single-prompt chat.

What if I just want to try Junior?

Start a free trial at /register — no credit card, 14 days, first workflow live in under 10 minutes.

Can I see which Claude model Junior used for a given task?

Yes — Junior's audit log records the model, tool calls, and tokens for every run, so you can reconcile cost and behavior per task.

Does Junior work without internet for the Claude API?

No — Junior is a cloud service and needs network access to call the model and your connected tools. There's no on-prem deployment today.

Home · Compare

Junior vs Claude

Claude is a powerful chat model you prompt; Junior is an AI employee that joins your Slack or Teams, remembers your team, and ships work without being asked.

Summary

Claude is Anthropic's flagship chat model — available via the Claude app, the Claude API, and the Claude Code CLI for developers. Junior is an AI employee that joins your Slack or Microsoft Teams workspace, remembers your team's context, and proactively delivers work without being asked. Junior supports Claude models and lets you select them on a per-AI-employee basis. Pick Claude if you're a developer or power-user happy to prompt; pick Junior if you want a coworker that ships outputs into the channels your team already uses.

Pick Junior if

Teams that want recurring work delivered into a shared channel — weekly reports, follow-ups, monitoring, cross-tool workflows — with no terminal or API skills required.

Pick Claude if

Developers and individual power-users who want a flagship chat model (via app, API, or Claude Code CLI) and are comfortable wiring up the rest of the workflow themselves.

Side-by-side capabilities

Junior vs Claude: capability comparison
Capability	Junior	Claude
Lives inside Slack or Microsoft Teams	Yes	No
Persistent memory of your team, docs, and history	Yes	Per-conversation; Projects on paid plans
Acts proactively without being prompted	Yes	No
Connects to Gmail, Notion, HubSpot, GitHub, ad platforms	3,000+ integrations	MCP + connectors (manual setup)
Scheduled tasks and recurring workflows	Yes	No
Review-first / approval-gated execution	Yes	No
No code or terminal required	Yes	Claude.ai app, yes; Claude Code CLI, no
Uses Claude under the hood	Yes — Anthropic Claude models are supported and selectable per AI employee	Always Claude
Audit log + per-AI-employee budget cap	Yes	No
Multiple AI employees with distinct roles	Yes	Projects ≈ folders, not separate workers

A chat model is not the same shape as a coworker

Claude (the app, the API, Claude Code in the terminal) is a model wrapped in three surfaces. Each one waits for a human prompt and returns an output. The unit of work is the request. Junior is shaped differently: it's a persistent worker with a name, a role, an org-scoped memory, a budget, and a calendar of recurring tasks. The unit of work is the deliverable. When a marketing manager hires "Maya" as their AI employee, Maya isn't a model session — she's an entity in Slack who remembers which Notion page she updated last Tuesday, which HubSpot deals she's tracking, and which channels her boss wants her to summarize each morning. Claude is excellent at being a model; Junior is built to be the entity wrapped around that model.

Persistent org memory vs per-conversation Projects

Claude's most team-shaped feature is Projects: a folder of documents and instructions you can scope a conversation to. It works, but the memory boundary is the project, and you have to think about which project a given task belongs in. Junior's memory boundary is the tenant — a Junior hired for your team knows everyone on the team, every connected tool, every prior workflow, and every approval decision the team has made. New conversations don't need re-onboarding; new channels Junior is invited to inherit the same memory. This matters for recurring work where Tuesday's report should know what Monday's report said.

When Claude is the right answer (and Junior uses Claude)

There's no antagonism between the two products. Junior is model-agnostic by design and supports Claude models — you can pick which Anthropic model each AI employee uses from the dashboard. The choice between Claude and Junior is really a choice between buying the raw model and buying the worker assembled around it. If your team's bottleneck is a developer who wants a sharper CLI and direct model access, Claude Code is the right tool. If the bottleneck is recurring cross-tool work that nobody on the team has time to maintain, the assembled worker — with memory, scheduling, approval gates, audit logs, and Slack/Teams presence — is what unblocks it.

When to choose which

Choose Junior when

You want recurring work shipped without prompting.
You want one entity that remembers the whole team's context.
You want approval gates and a per-employee budget cap.
Your team works in Slack or Microsoft Teams, not a terminal.
You're fine letting Junior pick the right Anthropic model per task.

Choose Claude when

You're a developer comfortable in Claude Code's CLI.
You only need a chat surface — no scheduling, no memory across sessions, no tool execution.
You want direct model access via API for custom orchestration.
You want to pin a specific Anthropic model version yourself.

FAQ

Does Junior use Claude?: Yes. Junior supports Anthropic Claude models, and you can select the model on a per-AI-employee basis from the dashboard.
How is Junior different from Claude Code?: Claude Code is a terminal-first developer tool — strong for developers comfortable in a CLI. Junior is a workspace coworker for whole teams — it lives in Slack or Teams, watches channels you point it at, and ships outputs without anyone opening a terminal.
Can I run Junior with my own Claude API key?: Pricing on /hire bundles model cost. Self-bring-your-own-key isn't supported today; if you need it, talk to a human via the demo form.
Is Junior faster than Claude for one-off questions?: No — for ad-hoc Q&A, opening claude.ai is faster. Junior is built for recurring, cross-tool workflows, not single-prompt chat.
What if I just want to try Junior?: Start a free trial at /register — no credit card, 14 days, first workflow live in under 10 minutes.
Can I see which Claude model Junior used for a given task?: Yes — Junior's audit log records the model, tool calls, and tokens for every run, so you can reconcile cost and behavior per task.
Does Junior work without internet for the Claude API?: No — Junior is a cloud service and needs network access to call the model and your connected tools. There's no on-prem deployment today.

Try Junior for your team.

Free trial · $100 credit. No credit card. Slack or Teams. First workflow live in 10 minutes.

Start free trial →Hire Junior