What Otter AI Actually Is
Otter.ai is an AI meeting assistant that joins your Zoom, Google Meet, and Microsoft Teams calls automatically, transcribes everything in real time, generates summaries with action items, and builds a searchable archive of every meeting you have ever attended. It was founded in 2016 and has been iterating on meeting transcription longer than most competitors in this space.
The 2026 v4.0 release introduced the Conversational Knowledge Engine — a system that treats your entire meeting history as a queryable database, letting you ask natural language questions across months of conversations and get cited answers back. This is the feature that most changes how meeting notes actually get used after the meeting ends.
Getting Set Up
Sign up at otter.ai with Google or email. The free Basic tier gives you 300 transcription minutes per month with a 30-minute cap per conversation — enough to evaluate whether the workflow fits your needs. Connect your calendar and Otter will offer to auto-join meetings, which you can toggle per meeting.
The iOS and Android apps handle in-person meetings and interviews when you are not at a desk. All transcripts sync to the cloud automatically and become searchable from any device.
How the Transcription Actually Works
OtterPilot joins your calendar meetings as a visible bot participant named Otter.ai Notetaker. During the call, a live transcript builds in real time with speaker identification — the system labels speakers automatically and improves accuracy as it learns voices over time.
After the meeting, Otter sends an email to attendees with a summary, action items, and an outline. The transcript is also available in the app with synchronized word highlighting so you can read along during the call. Speaker labels can drift in long calls with many similar voices; adding custom vocabulary under Settings for names, product terms, and acronyms noticeably improves accuracy.
The Conversational Knowledge Engine
The feature that changes post-meeting utility: AI Chat lets you ask questions across your entire meeting history. What did I say to the client about the timeline? What decisions were made in the board meeting last quarter? Which meetings mentioned our competitor? The answers come back with citations to the exact transcript lines, letting you verify and drill deeper.
For teams, the Business plan adds collaborative editing — team members can highlight, comment, and add photos to transcripts in real time, turning meeting notes into shared documents rather than personal archives.
Integrations and Workflows
Otter connects to Slack for automatic transcript posting, Salesforce and HubSpot for CRM logging, and Zapier for custom workflow automation. The MCP server — added in 2026 — lets AI tools like ChatGPT and Claude query your meeting library directly as a data source, which is useful for preparing briefs before meetings or synthesizing follow-up actions across multiple recent calls.
The key limitation worth knowing: Otter joins calls as a visible bot, which some meeting participants find disruptive. For client-facing or sensitive meetings, you can decline the bot attendance or use the mobile app to record and transcribe after the fact.
The Free Tier and When to Upgrade
The Basic free plan at 300 minutes per month covers occasional use adequately. Pro at roughly $8.33 per month billed annually is the realistic floor for daily users — 1,200 minutes per month with 90-minute meeting caps, advanced search, and custom vocabulary. Business at $20 per user per month billed annually adds unlimited meetings, 4-hour meeting caps, admin analytics, and CRM integrations. Enterprise adds SSO, HIPAA compliance, and a dedicated account manager for regulated industries.
built on Visual Studio Code, acquired by SpaceX in August 2026. It embeds AI agents directly into the editing workflow — the AI can read your codebase, suggest changes, write code, run commands, and manage multi-file refactors without you switching tools. If you write code and have not tried an AI code editor yet, Cursor is one of the most capable starting points.
Cursor 3.0, released October 10, introduced a major architectural shift called Projects — a coordination agent system that manages long-running, multi-agent workflows across months of work. The basic editing experience is free to start with; the Teams plan at $30 per user unlocks the full agent workspace.
Setting Up for the First Time
Download Cursor from cursor.com and sign in with email, GitHub, or Google. The editor opens like a standard VS Code environment — if you have used VS Code before, the transition is seamless. The key difference: press Cmd+I to open the Agent panel, which is where most of the AI interaction happens.
For the free tier, you get Composer — Cursor own model — with monthly usage limits. The free tier is enough to evaluate the core experience and decide if the paid plans are worth it for your workflow.
The Agents Window: What Changed in 3.0
Cursor 3.0 introduced the Agents Window — a dedicated space for running multiple AI agents in parallel. Rather than one agent running in a single chat, you can now launch multiple Agent Tabs side by side or in a grid, each working on different tasks simultaneously. Agents can run on your local machine, in isolated git worktrees, on cloud instances, or over SSH.
The key practical benefit: you can have one agent implementing a feature while another runs tests and a third reviews the pull request — all in the same interface without switching context.
The Basic Editing Loop
The simplest way to use Cursor: open a file, press Cmd+K for inline autocomplete suggestions, or Cmd+I to describe what you want in natural language and watch the editor apply changes across your project. The AI reads surrounding code to understand context, which makes suggestions more accurate than copy-pasting into a general chatbot.
For refactoring: describe the change you want across your codebase and Cursor will show you a diff before applying anything. You review and approve each change, which keeps you in control even for large multi-file modifications.
Projects: Managing Long-Running Work
Projects, now in beta and rolling out from October, is Cursor most significant capability expansion. A Project is a coordination agent that maintains context across months of work, can delegate to thousands of sub-agents, and can run on schedules without prompting. It keeps a shared context across all agents — files, research notes, and knowledge about the codebase — that compounds over time.
The coordinator does not write code itself; it plans work, delegates to specialized agents, and collects completed work for your review. When you close your laptop, the project keeps running on a cloud machine. When a task needs testing locally, it spawns a local agent on your machine.
Pricing: What to Start With
Existing subscribers keep their legacy plans: Pro at $20 per month includes Claude and GPT usage. New sign-ups see The X Subscription: Free, Plus at $8 per month with Grok models only, Premium at $30 with generous limits, Super at $100, and Ultra at $200. On Premium and above, Claude and GPT are billed on top at API list prices.
If you are starting fresh and want the full experience, Premium at $30 is the most practical plan — it carries what the old Pro plan offered. If you just want to try it, the free tier is genuinely useful for evaluating whether AI-assisted editing fits your workflow.