Hi guys,
I recently read an amazing article about hyper-productivity, which is changing my whole mindset around AI.
Talking to ChatGPT and saving a few minutes here and there is cool.
But, to keep up in the new era, we must leverage the compounding benefits of AI by building growth engines that learn, evolve, and iterate on themselves.
We need to stop laying the bricks and instead become architects.
You’ll learn more about hyper-productivity in the “Content to enjoy” section.
The theme of this week’s guide is Gemini 3 Pro. The insanely powerful model is shaking things up in the AI world and proves that Google is well and truly back in the game.
I’ll go over the key capabilities that set Gemini 3 Pro apart, and you’ll get 10 fantastic business-focused use cases you can try today.
The next AI Operator live session will be on Friday December 5th, and it’ll be Zapier 101.
Sign up now so you don’t miss out!
Today at a glance:
Google releases Nano Banana Pro and Anthropic launches Claude Opus 4.5, a "deep-thinking" coding powerhouse.
In today’s guide, you’ll learn how to leverage Google’s latest model with 10 unbelievable Gemini 3 Pro workflows.
Hyperproductivity: Steve Newman’s article on "Hyperproductivity" explores the future of AI-powered work. He interviews an engineer using complex, autonomous AI systems to ship dozens of products.
Tools of the week include Nano Banana Pro, Google AI Studio, Brain.fm and Magic Path.
News:
Nano Banana Pro: the most useful image generation tool for business
The recently-launched Nano Banana Pro has already established itself as the most useful image generation tool for business today.
It’s easy and fun to generate marketing content like infographics, carousels, and slide decks—with the unique ability to include correct, legible writing rather than the usual gibberish found in AI images.
I used it to make the banner for this newsletter, and I can’t wait to experiment more!
Anthropic launches Claude Opus 4.5
The battle for the "best model in the world" just intensified. After Google’s blockbuster Gemini 3 Pro release earlier this week, Anthropic has unveiled Claude Opus 4.5, aiming to reclaim the throne for high-level intelligence and coding.
The "Effort" Dial — A new parameter allows developers to control thinking depth—balancing speed against complex reasoning to match competitors.
Agentic Supremacy — With enhanced coding, agents and computer use tools it targets the one area where Gemini was pulling ahead: autonomous agent workflows.
Aggressive Pricing — ****At $5/million input, Anthropic is undercutting the market to lure enterprise users back from Google and OpenAI.
The model wars are heating up, and I’m excited to see who makes the next move (over to you, Sam...)
The Guide:
10 High-Leverage Ways to Use Gemini 3 Pro Today
Gemini 3 Pro is a huge milestone in the evolution of AI, marking a transition to agentic reasoning.
This model can pause, plan, test hypotheses, and self-correct before it ever sends you a response.
It’s been called the most powerful model in the world, so what can you do with it?
This guide will give you 10 powerful Gemini 3 Pro use cases to try immediately.
The new rules of the game
First, you have to shift your mindset and update how you engage with Gemini 3.
Rule #1: Treat it like a junior engineer
Gemini 3 Pro can reason and "one-shot" (build in a single prompt) complex apps. But to do this successfully, you need “architectural prompts, not vague ones.
Don't say: "Build me a website."
Do say: "Build a React application with Tailwind CSS, use BEM naming conventions, and ensure semantic HTML."
With strict constraints, the model can act more like an engineer.
Rule #2: Understand "Deep Think"
This is the engine under the hood. When you enable "Deep Think", the model generates hidden "thinking tokens" to map out a solution.
The strategy: Use the “thinking level” parameter. Set it to "low" for quick emails. Save "high" for complex coding, financial modelling, or data analysis where accuracy matters more than speed.
Rule #3: Think beyond text output
Gemini 3 Pro introduces Generative UI which can build tools for you in real-time. If you ask for a loan comparison, instead of just writing a paragraph, it will code and render an interactive mortgage calculator with working sliders directly in your chat window.
10 high-leverage use cases for founders
Here are 10 specific workflows to save time, money, and mental energy starting today.
Operational and strategic efficiency
1. The Contract Reviewer
Stop paying hourly legal fees for initial reviews. With the 1 Million Token Context Window and Deep Think, you can upload your entire shareholder agreement or a 50-page vendor contract.
The prompt: "Review this contract against our standard terms (uploaded). Simulate a 'hostile' negotiation and identify the 3 clauses that expose us to the most liability. Think step-by-step."
Why it works: It reasons through legal logic and potential future scenarios.
2. The "Roast My Deck" Simulator
Before you pitch to VCs, let Gemini destroy your logic.
The workflow: Upload your pitch deck PDF. Enable Deep Think (High). Ask it to adopt the persona of a skeptical Series A investor.
The output: It will break down your market sizing logic, challenge your CAC/LTV assumptions, and find holes in your narrative before you enter the boardroom.
3. The "Shadow" Co-Founder (strategic debate)
Use Thought Signatures to maintain a weeks-long strategic debate without restarting the context every time.
The workflow: Keep a running thread on "Q4 Pivot Strategy." Because the model encrypts and remembers its "train of thought" from previous turns, you can bounce complex ideas back and forth over days. It remembers why you rejected the pricing model last Tuesday.
4. The 24/7 "Code Janitor"
Use Google Antigravity to handle the tech debt you’ve been leaving in the backburner.
The workflow: Assign an asynchronous agent to "write unit tests for the entire API" or "update all dependencies" while you sleep.
The result: The agent browses files, runs terminal commands, and produces an "Artifact" (like a diff or plan) for you to approve in the morning.
Sales and product growth
5. The Dynamic Sales Closer
Stop sending static PDF pricing sheets. Use Generative UI during sales conversations.
The workflow: When a lead asks about enterprise volume discounts, prompt Gemini to "Build a sliding-scale pricing calculator based on our tiers."
The result: It renders an interactive slider in the chat. You (or the client) can drag the slider to see exact costs instantly, speeding up the "yes".
6. The "Vibe" Prototyper
You have an idea for a feature but no design skills. Use "Vibe Coding".
The workflow: Sketch a rough layout on paper. Upload a photo of it. Tell Gemini: "Build this dashboard. Make it look like a futuristic, dark-mode nebula."
The result: It translates your abstract aesthetic description into actual React/Tailwind code with effects and animations.
7. Automated User Testing
You don't need to manually click through your app to check for bugs.
The workflow: In Google Antigravity, assign an agent to "Test the user signup flow".
The result: The agent produces a screen recording (Artifact) of itself interacting with your app, showing you exactly where the UI breaks or where the flow gets stuck.
Marketing and brand
8. The Brand Consistency Enforcer
Generating AI images usually results in random faces. Nano Banana Pro fixes this.
The workflow: Upload up to 14 reference images of your product or brand mascot.
The result: Generate infinite social media assets (posters, scenes) where the character/product looks identical in every shot, ensuring brand consistency across your campaign.
9. The "Podcast" Miner
You don't have time to listen to 10 hours of competitor podcasts or industry panels.
The workflow: Upload the raw video or audio files directly (Native Multimodality).
The prompt: "Listen to these 3 hours of [Competitor] interviews. Extract their roadmap hints and any mention of their pricing strategy."
Why it works: It processes audio natively without needing a transcript first, catching nuances in tone.
10. Legible Marketing Mockups
Finally, you can create mockups that don't have gibberish text.
The workflow: Ask for "A billboard in Times Square displaying the text 'LAUNCHING TODAY'."
The result: Unlike older models, Gemini 3 Pro renders accurate, legible text on objects, letting you create realistic marketing mockups for your landing page in seconds.
The drawback: lazy research
While Gemini 3 Pro is a genius at reasoning, it’s an intern at research. Early tests show the model often performs a quick, surface-level Google search instead of the deep dive you need. In some cases, it even invents reasons to stop searching early just to finish the task.
My advice: Trust Gemini 3 Pro to analyze the data you give it (like uploaded PDFs or spreadsheets).
Don’t trust it to browse the web for complex due diligence. Always verify external sources.
Content to enjoy:
Article: Hyperproductivity: The Next Stage of AI? by Steve Newman
Hyperproductivity is productivity as a compounding loop: tools that build better tools and agents that update their own skills, with humans serving as the orchestrators.
In this article, Steve Newman profiles engineer Liu Xiaopai, the poster-child of hyperproductivity. Xiaopai explains how he ships dozens of profitable products almost single-handedly by automating their entire lifecycles. He contrasts his achievements with the exhausting demands of the work.
If you want to understand the trajectory of AI-powered work (and what might soon be possible for you), start here.
Tools to try:
Nano Banana Pro — Generate useful images and visual content in hundreds of different formats.
Google AI Studio — An online workspace to test, build, and prototype with Gemini models. Create AI-powered prompts, chatbots, and apps without needing deep technical setup.
Magic Path — An AI-powered design tool that lets you describe what you want in plain English (or via reference images), and it generates user-interfaces — websites, app screens, dashboards — on an “infinite canvas.”
TL;DR:
News — Competition heats up as Google releases Nano Banana Pro and Anthropic launches Claude Opus 4.5, a "deep-thinking" coding powerhouse.
The Guide — We break down 10 high-leverage Gemini 3 Pro workflows, from "one-shot" app building to "roasting" your pitch deck.
Content — Steve Newman’s article on "Hyperproductivity" explores the future of AI-powered work and how engineers are shipping dozens of products single-handedly.
Tools — Nano Banana Pro for visuals, Google AI Studio for prototyping, Brain.fm for focus, and Magic Path for instant UI generation.