Paul's course knowledge base

Gemini

First met in aicp-c3-basecamp-01 · Chat assistants · Models & model access

HomeBrowseContentsConceptsToolsSpeakersFiguresQuotesConcept mapFindFindSearchTranscriptsStudyRefresherPart 107 studyCertificatesSEO/GEOOut of date?MineProjectsAboutAdmin

Google's assistant and model family — huge context, deep Google-app gravity, and gems as saved custom assistants.

Pick this when

You need a very large context window cheaply, or you're already living in Google Workspace.

What makes it different

Context windows around 1M tokens, which makes whole-document work practical

Gems: saved custom assistants, the natural home for a meta prompt in Google's world

Gemini 3 Flash proved the cheap-and-fast option for voice agents in BC6

What it does

Answers with very large context, including big uploaded documents

Stores reusable instruction sets as gems

Powers Google AI Studio, Antigravity and the Flash models used for voice

What it does not do

Common wrong expectations

Doesn't have connectors — as of BC5, none at all, while Claude and ChatGPT both shipped them

Where it sits in a stack

The large-context and Google-ecosystem seat; increasingly the cost-effective model choice for latency-sensitive work.

Shelf

Chat assistants · Models & model access

Our status · not-using

No gap the Claude stack doesn't cover.

For your projects

Google AI Studio's free 1M context is the practical route if a future task needs a whole transcript in one window without spending Fable budget.

Freshness

current

Related tools

builds onGoogle AI Studiocompetes withChatGPTcomplementsAntigravity

google-ai-studio: the free playground surface for Gemini

chatgpt: peer assistant

antigravity: Gemini-driven coding agent

Every moment the course touched it (35)

SessionCoverageMomentWhat happened
Basecamp 6: Voice Agentsdemonstrated1:09:54His default voice LLM — ~half GPT-4.1's cost, faster, visibly dropped cost and latency on selection
Session 3: Building Elegant Landing Pagesdemonstrated1:06:49Video generation for the hero workflow (free tier carried the demo); AI Studio + API key removes the watermark; video model heard as 'Omni' (unverified)
Google Colab with Gemini: One Prompt to a Full Python Notebook, and Fixing a KeyErrordemonstratedPlan and auto-run
Gemini Overview: Multimodal Uploads, @-Tagging Google Apps, Saved Info Personalisation, and Honest Weaknessesdemonstrated
Gemini Gems: Saved Custom Assistants with Instructions and Knowledge Files, and Choosing Flash or Pro with Show Thinkingdemonstrated
Gemini Deep Research and Multimodal Analysis: Role-Task-Constraint Prompts, Editing the Research Plan, Export to Docs, and a Competitor Posterdemonstrated
Gemini Canvas and Google Workspace Apps: Infographics and Docs in a Side Panel, Tone Slider, Export to Docs, and Calendar and Gmail Queriesdemonstrated
AI Career Coach: a Custom GPT That Knows Your Resume, and a Gemini Gem Industry ResearcherdemonstratedGems
Day 1: The Magic Prompt Formula, a Toolkit Tour, Thinker-and-Doer Data Analysis, the Blind-Men-and-the-Elephant Research Method, and a Living Pitch DeckdemonstratedMagic prompt demo; Gmail sidebar
Day 2: Custom GPTs and Gems, Markdown Prompting, Cloning Your Writing Style, MCP in Daily Use, a Vibe-Coded Calorie Counter, and the Consultant-to-SaaS RoaddemonstratedGems; mentor bot with Knowledge
Evaluating AI Research Tools, Part 1: the Bake-off Method, a Market-Landscape Report, and a Notion Competitor BenchmarkdemonstratedDeep Research
Evaluating AI Research Tools, Part 2: Buyer Personas, Marketing Trends, a Content-Gap Analysis, the Verdict, and Adapting a Proven PromptdemonstratedDeep Research - first choice
AI Masterclass: Exploring Antigravity 2 — Four Products, Four Slash Commands, and Two Live Buildsdemonstrated3.5 Flash as the launch model — speed positioning, quota tiers, bundled Claude access
Office Hours VIII (platform label 'XIII'): App-Store Economics, Vetting a Co-Founder Without Your Code, Android Accessibility Agents, and OpenClaw vs HermesexplainedFlash as the fast routing model
Introduction to AI-Powered Content Creation: the Five-Step Stack, the 70/30 Split, Tool Roles, and a Starter Challengeexplained
Prompt Engineering for Writers, Part 1: Weak vs Professional Prompts, COSTAR, a Five-Email Worked Example, the Meta-Prompt, and Three Prompting ModesexplainedGems
Basecamp 1: Prompting & RAGsmentioned2:32:15Custom gems as meta-prompt containers
Basecamp 5: MCPmentioned2:25:09No connectors yet at session time
Office Hours I: Vetting an Idea, Niching Against the Frontier Labs, and Paul's Aircraft-Parts App (PWA, and 'What If I Get Hit by a Bus')mentionedDeep research for idea vetting
Session 11: Overview of Claude Code & Key Concepts (Installs, Modes, Context Files, Skills, Hooks & MCPs)mentioned2:13:03Cost alternative — 'even on free tier'
Session 20: Complete LinkedIn Agency Automation — Carousels + Newsletter from One Content SystemmentionedDharmesh's admired UI turned out to be Gemini output, no skill involved
Claude Overview: the Long Context Window, Plan-Gated Models, Changing Its Tone with Styles and Preferences, and Projectsmentioned
Pre-Launch Design Testing with an AI Review Agent: Audits, Copy Review, A/B Comparison, Fix Loop into Claude Code, and Ad/Video Scoringmentioned
Google Search Ads Fundamentals: Ad Rank and Quality Score, Extensions, Account Structure, Match Types and Smart Biddingmentioned
Learning Topics: Sessions 1–5 — Foundations to AI DirectormentionedNamed among assistants; Gems as the saved-assistant equivalent (limited in source snapshot); Nano Banana image family in session 5
Genspark AI Slides and Model Selector: Editable Decks with Fact-Check and Export, and Picking the Right Model in AI Chatmentioned
AI Product Ads: Turn a Phone Photo into an Animated Poster in Canva Magic Studio, then a Video Shot in Gensparkmentioned
NotebookLM Overview: a Source-Grounded Research Assistant, Notes Formats, Citations and a Start-Small Workflowmentioned
Building a Notion Database with AI: Pages with Properties, "Build with AI", and Kanban and Calendar Viewsmentioned
Perplexity Deep Research, Citations and the Model Selector: Three Search Tiers, Back-Checking Every Line, and Choosing a Modelmentioned
AI Sprint: GPT-6 Astra Live - Day 2 (Restaurant-Chef Analogy, Model-Harness-Workflow, the Six-Block Delegation Brief, a Ten-Page Launch Pack in One Loop, and Making Astra Argue Against Itself)mentionedDismissed as not in the race
AI Sprint: Choosing the Right Problem — Day 2 (A Nine-Step Worksheet from Pain Inventory to PRD, and the Honest Screen for Whether AI Belongs at All)mentionedSame - named, then banned for the pain-inventory step
AI Sprint: Build Your Own AI Skills - Day 1 (Solve It Three Times Before You Build the Skill, Prompt / Skill / Project Rule, the Video-Restoration Walkthrough, and the Two-Skill Voice-Then-Deslop Pipeline)mentionedNot yet at skills parity
SuperJoin Deep Dive: AI Data Analysis Inside Google Sheets - Connectors, Prompt-Driven Reports, and In-Cell SuperGPT Formulasmentioned
Weekly AI Updates - October 2026 digest (Oct 7 so far: Gemini 4 'Argon', America.gov AI, Meta Muse, NVIDIA OpenShell, Grok Bot rewards, Copilot teammates, Claude Sonnet 5.5 deep dive; Mark Palmer's GBox showcase; Oct 14 so far: Google Health Coach, the White House frontier-AI accord, OpenAI Dev Day 2026, Anthropic's Oct 8 usage-policy update, Hark, Microsoft MAI voice models, Claude Haiku 5.5)mentioned'Gemini 4 Argon' as heard

← All tools by category