Think Alongside AI

A chatbot gives you one answer. You take it or you don't. If you want to explore a different direction, you either start over or scroll up and try to steer the conversation back. Your previous ideas get buried. Your good work gets lost in the scroll.

TeamTeacher is built for the way teachers actually plan — iteratively, in branches, building on what's working while exploring what might work better. You don't just prompt the AI and hope. You think through problems together.

Explore Without Losing Your Work

Every AI message in TeamTeacher has a fork button. Click it and you get a new conversation that branches from that exact point — carrying all the context, documents, and notes from the original, but free to go in a completely different direction.

This sounds small. It changes everything.

Develop a unit in branches. Start a conversation to plan a unit on ecosystems — learning objectives, essential questions, the arc of the unit. Once the foundation is solid, fork from there. One branch develops Lesson 1 in detail. Another branch builds the lab investigation. A third designs the summative assessment. Each branch has the full unit context, so everything stays aligned — but each one can focus on exactly what it needs to.

Try different approaches without risk. You've got a good lesson outline but you're not sure about the scaffolding. Fork. Try one approach in one branch, a different approach in the other. Keep whichever works. The original is untouched.

Differentiate from a common plan. Fork your lesson plan three ways: one for students who need more support, one for your advanced learners, one with ELL accommodations. All three share the same core content and objectives. Each one is tailored.

The forked conversation links back to the original, so you can always trace where a branch started and navigate between related conversations.

Keep Track of What Matters

As you work through a conversation, ideas accumulate. Key decisions get made on line 47 of a long thread and then you can't find them later. TeamTeacher's conversation notes solve this.

Notes are a space within each conversation — visible to both you and your agent — where important information gets captured as you go. Your agent can add to the notes as decisions are made ("We agreed on three learning objectives: ..."), and you can edit them directly whenever you want.

When you fork a conversation, the notes come with it. So if you spent thirty minutes planning a unit and captured the key decisions in notes, every branch of that work starts with those decisions already documented. No re-explaining, no scrolling back to find what you agreed on.

Your Agent Remembers Previous Work

Sometimes the context you need isn't in the current conversation — it's in something you worked on last week. Your agents can search across your past conversations to find relevant work and pull in what's useful.

Planning a new chemistry unit? Your agent can look up the unit you planned last semester and summarize the approach you took — what worked, what the learning objectives were, how you structured the assessment. You're building on your own past thinking instead of starting from a blank page every time.

This works alongside your document library, where finished materials live. Conversations hold the thinking process; documents hold the polished output. Your agents can draw on both.

What This Looks Like in Practice

Here's a realistic planning session:

You sit down on Sunday evening to plan next week's MYP Science unit on chemical reactions. You open a conversation with Minerva in your "MYP Science Year 4" folder — she already has access to your unit overview and previous lessons.

You start talking through the week: what concepts to cover, how to sequence the activities, where students struggled last time. Minerva takes notes on the key decisions as you go. She pulls up your previous lesson on atomic structure to check what background knowledge students already have.

Once the week's plan feels solid, you fork. In one branch, you develop Tuesday's lesson in detail — the hook, the demonstration, the practice activity. In another branch, you design Thursday's formative assessment. Minerva consults the MYP knowledge base to make sure the assessment aligns with Criterion C strand descriptors at the right achievement level.

Both branches share the weekly plan as context. Both build on the notes you captured. You end up with a coherent week instead of disconnected pieces — and the whole conversation history is there if you want to revisit your reasoning.

Next week, you'll fork from the main plan again for Week 2. The context keeps building.

Start a conversation →