How to use these cards: Print one per page, laminate, and keep at your desk. Each card covers one tool — what it is, when to use it, and how to avoid the most common mistakes.
1Card 1 — Gemini¶
gemini.google.com · Personal Google account required
What it is: Gemini is Google’s conversational AI assistant — a general-purpose thinking partner that reads, writes, analyzes, and reasons across text, images, and documents.
1.1Best For¶
| # | Teacher Use Case |
|---|---|
| 1 | Drafting lesson plans, rubrics, and syllabi from a brief description |
| 2 | Differentiating a single activity for multiple reading levels |
| 3 | Generating quiz questions, discussion prompts, or exit tickets |
| 4 | Summarizing research articles or PDFs you upload |
| 5 | Writing parent/guardian communication in plain language |
| 6 | Brainstorming project ideas, analogies, or real-world examples |
1.2Access¶
Go to gemini.google.com
Sign in with your personal Google account
School/institution accounts may have Gemini features restricted by IT policy — use personal account for full access
1.3Gems — Your Custom AI Personas¶
What they are: Gems are saved Gemini configurations with a custom name, instructions, and persona. Create one for each role you play repeatedly (e.g., “Rubric Writer,” “Quiz Generator,” “Parent Email Helper”).
How to create a Gem (3 steps):
Click “Explore Gems” → “New Gem” in the left sidebar
Give it a name, describe its role, and paste in your standing instructions (e.g., “You help me write rubrics aligned to Bloom’s Taxonomy for undergraduate business courses”)
Click Save — the Gem appears in your sidebar for instant reuse
1.4Power Tips¶
1.5Watch Out For¶
2Card 2 — NotebookLM¶
notebooklm
What it is: NotebookLM is an AI research assistant that reads only the sources you provide — letting you query, summarize, and explore your own documents, articles, and videos without the AI wandering off into the open internet.
2.1Best For¶
| # | Teacher Use Case |
|---|---|
| 1 | Building a private Q&A assistant from your course readings or textbook chapters |
| 2 | Generating a Study Guide or FAQ from your syllabus and lecture notes |
| 3 | Creating an Audio Overview (podcast-style summary) students can listen to |
| 4 | Cross-referencing multiple sources to find connections or contradictions |
| 5 | Letting students explore a curated source set without access to unvetted web content |
| 6 | Reviewing and summarizing student-submitted papers or research packets |
2.2Access¶
Go to notebooklm
.google .com Sign in with your personal Google account
Create a new notebook for each project, course, or unit
2.3Key Features¶
| Feature | What It Does |
|---|---|
| Sources panel | Left sidebar — add, view, and manage all your loaded documents |
| Chat | Ask questions; answers cite the exact source and page number |
| Study Guide | Auto-generates key themes, glossary, practice questions, and timeline |
| Audio Overview | Produces a ~10-minute AI-hosted “podcast” summarizing your sources |
| Mind Map | Visual concept map showing how ideas connect across your sources |
2.45 Ways to Load Sources¶
URL — Paste any public webpage link (article, blog, news story)
PDF upload — Upload from your computer (research papers, handouts, textbooks)
Google Drive — Connect directly to Docs, Slides, or PDFs in your Drive
Paste text — Copy-paste any raw text directly into the source panel
YouTube link — Paste a YouTube URL; NotebookLM reads the transcript
2.5Power Tips¶
2.6Watch Out For¶
3Card 3 — AI Studio¶
aistudio.google.com · Free tier available
What it is: AI Studio is Google’s developer-facing playground for Gemini models — giving educators direct access to raw model settings, system instructions, and API keys to build custom AI-powered tools and workflows.
3.1Best For¶
| # | Teacher Use Case |
|---|---|
| 1 | Building a reusable AI tool with a fixed persona (e.g., a Socratic tutor that never gives direct answers) |
| 2 | Stress-testing prompts before deploying them in a class activity |
| 3 | Experimenting with multimodal input — upload images, audio, or video for analysis |
| 4 | Generating structured output (JSON, tables, formatted lists) for gradebook or LMS import |
| 5 | Getting an API key to connect Gemini to Google Sheets, Docs, or a course app |
| 6 | Comparing how different Gemini models respond to the same prompt |
3.2Access¶
Go to aistudio.google.com
Sign in with a Google account — free tier available (rate limits apply)
Paid tiers unlock higher quotas and access to the latest models
3.3Key Controls¶
| Control | What It Does |
|---|---|
| Model selector | Switch between Gemini Flash (fast/cheap), Gemini Pro (balanced), and Gemini Ultra (most capable) |
| Temperature | 0 = precise and deterministic → 1 = creative and varied. Use 0–0.3 for factual tasks; 0.7–1.0 for creative tasks |
| Token limit | Sets the maximum length of the AI’s response. Increase for long documents; decrease to force brevity |
| System Instructions panel | The hidden “rulebook” — write standing instructions the model follows for every message in the session |
3.4Power Tips¶
3.5Watch Out For¶
4Card 4 — Google Antigravity¶
antigravity.google
What it is: Google Antigravity is an agentic AI platform that lets you build and run autonomous AI agents — programs that can plan multi-step tasks, use tools (search, email, calendar, code), and remember context across long workflows without you manually directing every step.
4.1Best For¶
| # | Teacher Use Case |
|---|---|
| 1 | Automating repetitive administrative workflows (e.g., weekly grade summary emails) |
| 2 | Building a research agent that searches, reads, and synthesizes sources on a topic |
| 3 | Creating a student-facing tutoring agent with guardrails and a fixed persona |
| 4 | Running multi-step course design: outline → lessons → assessments → rubrics in one run |
| 5 | Connecting your course tools (calendar, Drive, Sheets) into a single orchestrated workflow |
| 6 | Prototyping AI-integrated classroom activities without writing code |
4.2Access¶
Go to antigravity.google
Sign in with your Google account
Some features may require joining a waitlist or early-access program — check the site for current availability
4.3Key Concepts¶
| Concept | Plain-English Explanation |
|---|---|
| Agent | The AI entity that executes your task — thinks, plans, and acts on your behalf |
| Goal | The high-level objective you give the agent (e.g., “Research three studies on active learning and write a summary”) |
| Tools | Capabilities the agent can use — web search, Google Drive, Gmail, code execution, etc. |
| Memory | Information the agent retains across steps or sessions — facts, prior outputs, user preferences |
| Guardrails | Rules and restrictions that keep the agent within safe, intended boundaries (e.g., “Only read files in this folder; never send emails without my approval”) |