StarZero All shows Try the podcast clipper →

Feb 15, 2025 · 29m · startup-ideas

Google AI studio replaces your AI tech stack (full demo)

Logan Kilpatrick · 21m spoken Greg Isenberg · 5m spoken
0:00 / 0:00
▶ Watch on YouTube →

gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions

Google AI Studio Lead Product Manager Logan Kilpatrick joins Greg Isenberg to demonstrate Gemini's advanced multimodal toolset—spanning massive video context windows, spatial reasoning, function calling, and real-time voice-vision collaboration—revealing how founders can rapidly build and launch AI startups.

How this conversation actually went

Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Greg holds 20.1% of the talking time here. How this is scored →

Greg as informed peer 3.8 Guest teaching 5.8 Guest disagreement 0.0 Greg pushing back 0.0
05100:0010:0020:002:28–6:15 · Greg as informed peer 4/10 Exploring AI Studio UI and Long-Context Video Analysis Logan introduces AI Studio's long-context capabilities by processing a 30-minute museum video, while Greg identifies business opportunities in data extraction for online directories.6:15–13:24 · Greg as informed peer 2/10 Sponsor Break: Startup Empire Membership Following a sponsor read, Logan demonstrates Gemini reasoning models and exposed UI thought traces, educating Greg on bringing free API keys into Cursor.13:25–16:46 · Greg as informed peer 4/10 Spatial Understanding and Computer Vision Business Applications Logan demos 2D bounding boxes and computer vision capabilities, answering Greg's prompt about business applications with examples in inventory and parking utilization.16:46–20:05 · Greg as informed peer 5/10 Native Function Calling and Rapid SaaS Prototyping Greg highlights how AI automation can replace expensive service agencies, while Logan demonstrates function calling and open-source GitHub starter templates.20:06–26:49 · Greg as informed peer 4/10 Multimodal Live API and Real-Time AI Co-Presence Demo Logan presents a live demo of the Multimodal Live API screen-debugging Python code in real time, leading Greg to reflect on the future of AI co-presence in the workplace.2:28–6:15 · Guest teaching 5/10 Exploring AI Studio UI and Long-Context Video Analysis Logan introduces AI Studio's long-context capabilities by processing a 30-minute museum video, while Greg identifies business opportunities in data extraction for online directories.6:15–13:24 · Guest teaching 6/10 Sponsor Break: Startup Empire Membership Following a sponsor read, Logan demonstrates Gemini reasoning models and exposed UI thought traces, educating Greg on bringing free API keys into Cursor.13:25–16:46 · Guest teaching 6/10 Spatial Understanding and Computer Vision Business Applications Logan demos 2D bounding boxes and computer vision capabilities, answering Greg's prompt about business applications with examples in inventory and parking utilization.16:46–20:05 · Guest teaching 5/10 Native Function Calling and Rapid SaaS Prototyping Greg highlights how AI automation can replace expensive service agencies, while Logan demonstrates function calling and open-source GitHub starter templates.20:06–26:49 · Guest teaching 7/10 Multimodal Live API and Real-Time AI Co-Presence Demo Logan presents a live demo of the Multimodal Live API screen-debugging Python code in real time, leading Greg to reflect on the future of AI co-presence in the workplace.2:28–6:15 · Guest disagreement 0/10 Exploring AI Studio UI and Long-Context Video Analysis Logan introduces AI Studio's long-context capabilities by processing a 30-minute museum video, while Greg identifies business opportunities in data extraction for online directories.6:15–13:24 · Guest disagreement 0/10 Sponsor Break: Startup Empire Membership Following a sponsor read, Logan demonstrates Gemini reasoning models and exposed UI thought traces, educating Greg on bringing free API keys into Cursor.13:25–16:46 · Guest disagreement 0/10 Spatial Understanding and Computer Vision Business Applications Logan demos 2D bounding boxes and computer vision capabilities, answering Greg's prompt about business applications with examples in inventory and parking utilization.16:46–20:05 · Guest disagreement 0/10 Native Function Calling and Rapid SaaS Prototyping Greg highlights how AI automation can replace expensive service agencies, while Logan demonstrates function calling and open-source GitHub starter templates.20:06–26:49 · Guest disagreement 0/10 Multimodal Live API and Real-Time AI Co-Presence Demo Logan presents a live demo of the Multimodal Live API screen-debugging Python code in real time, leading Greg to reflect on the future of AI co-presence in the workplace.2:28–6:15 · Greg pushing back 0/10 Exploring AI Studio UI and Long-Context Video Analysis Logan introduces AI Studio's long-context capabilities by processing a 30-minute museum video, while Greg identifies business opportunities in data extraction for online directories.6:15–13:24 · Greg pushing back 0/10 Sponsor Break: Startup Empire Membership Following a sponsor read, Logan demonstrates Gemini reasoning models and exposed UI thought traces, educating Greg on bringing free API keys into Cursor.13:25–16:46 · Greg pushing back 0/10 Spatial Understanding and Computer Vision Business Applications Logan demos 2D bounding boxes and computer vision capabilities, answering Greg's prompt about business applications with examples in inventory and parking utilization.16:46–20:05 · Greg pushing back 0/10 Native Function Calling and Rapid SaaS Prototyping Greg highlights how AI automation can replace expensive service agencies, while Logan demonstrates function calling and open-source GitHub starter templates.20:06–26:49 · Greg pushing back 0/10 Multimodal Live API and Real-Time AI Co-Presence Demo Logan presents a live demo of the Multimodal Live API screen-debugging Python code in real time, leading Greg to reflect on the future of AI co-presence in the workplace.

speaking balance: gold is Greg, purple is the guest (3 minute bins)

0:00 · Greg 48.8% · guest 51.2%0:00 · Greg 48.8% · guest 51.2%3:00 · Greg 14.5% · guest 85.5%3:00 · Greg 14.5% · guest 85.5%6:00 · Greg 34.1% · guest 65.9%6:00 · Greg 34.1% · guest 65.9%9:00 · Greg 5.4% · guest 94.6%9:00 · Greg 5.4% · guest 94.6%12:00 · Greg 0.5% · guest 99.5%12:00 · Greg 0.5% · guest 99.5%15:00 · Greg 34.2% · guest 65.8%15:00 · Greg 34.2% · guest 65.8%18:00 · Greg 0.9% · guest 99.1%18:00 · Greg 0.9% · guest 99.1%21:00 · Greg 0% · guest 100%21:00 · Greg 0% · guest 100%24:00 · Greg 41.2% · guest 58.8%24:00 · Greg 41.2% · guest 58.8%27:00 · Greg 21.1% · guest 78.9%27:00 · Greg 21.1% · guest 78.9%
Sharpest disagreement ▶ 23:12 Addressing live model rough edges and conversational flow limitations

In a completely collaborative episode, Logan offers self-critical pushback on his own demo, pointing out that conversational cadence and interruption timing still have rough spots.

 Hardest push from Greg ▶ 26:32 Defending early AI demos against perfectionist critics

Greg proactively rejects potential listener skepticism regarding voice intonations, arguing that even imperfect tooling sparks transformative product ideas.

 Biggest teaching moment ▶ 21:12 Live multimodal AI audio and visual debugging demo

Logan and Gemini AI Voice demonstrate real-time visual code inspection and error correction directly from a shared desktop screen.

 Greg holds their own ▶ 16:45 Framing agency automation and multimillion-dollar service replacement

Greg demonstrates his startup expertise by outlining how multimodal vision can displace high-overhead agency consulting tasks into automated software businesses.
the scores for every segment, with the reasoning behind each
ChapterTopicGreg as informed peerGuest teachingGuest disagreementGreg pushing backWhy
Exploring AI Studio UI and Long-Context Video Analysis 4500 Logan introduces AI Studio's long-context capabilities by processing a 30-minute museum video, while Greg identifies business opportunities in data extraction for online directories.
Sponsor Break: Startup Empire Membership 2600 Following a sponsor read, Logan demonstrates Gemini reasoning models and exposed UI thought traces, educating Greg on bringing free API keys into Cursor.
Spatial Understanding and Computer Vision Business Applications 4600 Logan demos 2D bounding boxes and computer vision capabilities, answering Greg's prompt about business applications with examples in inventory and parking utilization.
Native Function Calling and Rapid SaaS Prototyping 5500 Greg highlights how AI automation can replace expensive service agencies, while Logan demonstrates function calling and open-source GitHub starter templates.
Multimodal Live API and Real-Time AI Co-Presence Demo 4700 Logan presents a live demo of the Multimodal Live API screen-debugging Python code in real time, leading Greg to reflect on the future of AI co-presence in the workplace.

Statements from this episode (9)

Assertion Supported
Kilpatrick: Google reasoning model is available for free to developers
“We have our reasoning model available for free for developers. You can use it in AI studio. You can go and get an API key and use it as well.”
Logan Kilpatrick Feb 15, 2025 ▶ 8:05
Assertion Supported
Kilpatrick: Google AI Studio UI exposes model thoughts, but API abstracts them
“So we're showing these thoughts in the UI. If you were using the API, you actually wouldn't see these thoughts. It's sort of abstracted behind the scenes, but we showcase it in the UI just to sort of give you an intuition of what's happening.”
Logan Kilpatrick Feb 15, 2025 ▶ 10:26
Assertion Supported
Kilpatrick: Gemini natively incorporates spatial understanding and object localization
“So this is a capability that is like Baked into the model itself, where it's like really able to deeply understand different objects and the ways in which they're sort of visually represented.”
Logan Kilpatrick Feb 15, 2025 ▶ 13:50
Disclosure
Kilpatrick: Inventory management companies use Gemini bounding boxes for capacity tracking
“I've worked with a bunch of companies that do like inventory management and like this type of bounding box example, like matters a lot for them. So that like, they can just literally like snap pictures or have real time video feeds to like, you know, how much …”
Logan Kilpatrick Feb 15, 2025 ▶ 15:53
Insight
Isenberg: AI automation will replace human-run agency and consulting tasks
“There's a lot of service based businesses that, you know, their agencies, consulting firms that do a service, a painful task for a business. You don't have to have humans do that anymore. You can now use technology to go and automate a lot of that stuff. And t…”
Greg Isenberg Feb 15, 2025 ▶ 16:49
Insight
Kilpatrick: Combining Existing APIs with AI Unlocks Viable SaaS Businesses
“What happens if you take five random different products, and you bring them all together, and you put AI in the middle? Like, that's actually, like, a real business all of a sudden. And the amount of work to do that is, like, actually pretty pretty small relat…”
Logan Kilpatrick Feb 15, 2025 ▶ 19:15
Prediction Not checkable as stated
Kilpatrick: Real-time multimodal models will become co-present pair programmers
“I think the future that we're going to see with the multimodal live API is really the ability for the model to or just like generally for models to be co-present in the experiences that you are. And like, you could imagine your IDE is the best experience of th…”
Logan Kilpatrick Feb 15, 2025 ▶ 23:11
Prediction Not checkable as stated
Isenberg: Real-time screen-sharing AI assistants are the future of work
“When I see this is the future of work. You're going to have It makes sense that you would have this AI partner, you know, see what you're seeing and provide value and intelligence to whatever it is you're working on at any given moment.”
Greg Isenberg Feb 15, 2025 ▶ 24:48
Assertion Supported
Kilpatrick: Google AI Studio Offers 1.5 Billion Free Tokens for Gemini
“The API keys actually by default are also free, so you can get, like, 1.5 billion tokens using Gemini across a bunch of different models for free today.”
Logan Kilpatrick Feb 15, 2025 ▶ 28:47
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 300 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.

Try the podcast clipper Learn more