Work Like a Team of Four: The New AI Reality for Interior Designers | AI for Interior Designers™

AI for Interior Designers™ Podcast

Work Like a Team of Four: The New AI Reality for Interior Designers

A comprehensive fall 2025 update on where AI actually is — imagery, app building, agentic tools, and the Perplexity Comet browser that has quadrupled what one person can do in a day.

Ep. 60 Solo Episode ✦ Jenna Gaidusek September 2025

This blog was written using AI as a recap from the recording, then edited by the author for accuracy and details.

Key Takeaways

AI imagery is genuinely there now. With Nano Banana and Gemini, you can place real products into a real room photo at near-photorealistic accuracy — and work in conversational segments rather than one massive prompt.
Custom app building is accessible without a developer. Jenna built a full design workflow app — concept boards, color palettes, paint matching, product sourcing, image generation, and animated walkthroughs — using Gemini and vibe coding.
Agentic AI is not coming — it is already here and already useful. The Perplexity Comet browser handles email, calendar, research, and task management as an active participant in your workday, not just an answer machine.
Floor plans still require human judgment. AI imagery looks great at first glance but does not replace the critical thinking involved in planning a space that functions in the real world.
AI does not replace you — it replaces the two-hour tasks that are not your actual job. The goal is to offload the tedious so you can spend more time on the parts of design you actually love.

The Short Answer

As of fall 2025, AI is genuinely useful for interior designers in three distinct ways: it creates client-ready imagery with your actual products in your actual spaces, it runs entire workflow sequences through custom-built apps, and it handles the administrative layer of your day through agentic tools that act on your behalf. Combined, these three capabilities are not replacing the designer — they are giving one person the output capacity of a four-person team.

Where AI Imagery Actually Is Right Now

The imagery capabilities that have been "almost there" for the past two years have crossed a meaningful threshold. Google Gemini's Nano Banana — the image editing tool now embedded directly in many Android camera rolls — maintains product accuracy in a way that earlier tools did not. You can take a photo of a rug, provide measurements and a room photo, and place that exact rug into the exact space with close to 99% visual accuracy of the actual product.

The key technique shift: work in segments. Jenna spent time in earlier AI tools giving every instruction at once — add the rug, change the paint, swap the pendant, rearrange the furniture — and getting results that were partially right and partially rogue. The approach that works now is iterative: make one or two changes, get the result, then continue from that image. Each small conversation yields more reliable output than one exhaustive prompt.

One thing that has not changed: floor plans still require a human. AI imagery can look spatially convincing without the furniture actually fitting. Professional floor plan tools like MyDoma Visualizer remain essential — the AI visual communicates the concept, the floor plan confirms the reality.

On video: Gemini VO3 and Midjourney are Jenna's current choices for room walkthrough animations. The key prompt constraint — "walk forward in the space and don't make up anything you can't see" — keeps the output from hallucinating furniture that was not in the original image. Sora, OpenAI's video tool, is not currently in her workflow.

The Integrated Design App: One Flow Instead of Ten Tools

The immediate trigger for building the app was a real client project. Jenna took on a new home project for a returning client who wanted to work together again, and she realized she was jumping between tools at every step of the process. So she built an app that does the whole sequence in one place.

The Design Workflow App — What It Does

Input Voice clips, notes, product selections, and space photos — all fed into the app at the start of the process.

Design Board The app generates a concept board from the collected inputs and the direction you have described.

Color Palette Extracts a color palette board from the design board automatically.

Paint Match Takes those colors and finds the closest Sherwin-Williams and Benjamin Moore matches. No separate tool needed.

Room Visual Places specific products into the actual room photo via Nano Banana, with a preset button for common prompts.

Animation One button creates a walkthrough animation from the room visual. Standard prompts are pre-loaded; you can add specifics if needed.

The client-facing result: a design concept board, a color board with actual paint names, a mockup of the real products in the real space, and a short animated walkthrough — all produced from one workflow without switching between platforms. The tools she used to pay for individually to accomplish different parts of this sequence have been cancelled.

A light beta version is available inside The DAIly membership. The full version — with expanded product sourcing and VR-style web walkthroughs — is in development.

Agentic AI and the Perplexity Comet Browser: Working Like a Team of Four

Agentic AI is the difference between asking your AI a question and telling your AI to go do something. Not "tell me what is on my calendar" — "go check my calendar, look for conflicts on these dates, and add this event with these details." The AI does not just answer. It acts.

Jenna has been using the Perplexity Comet browser as her primary browser for nearly two months at the time of recording, having received early access. It has replaced both Chrome and Edge in her daily workflow. Here is what that actually looks like:

→

Start the day by asking it to summarize overnight emails and surface action items. It accesses the inbox directly because she has logged in via the browser.

→

Voice-activate the assistant while reading emails to draft responses, add calendar events, or send links to the right people — hands-free while doing other work.

→

Ask it to research a technical problem she was stuck on while building an app. It took over the screen, walked through the steps, and showed her the solution in two hours instead of a paid developer call.

→

Consolidate post-presentation cleanup — gathering all the links from a presentation, organizing them by recipient, and generating a Google Doc to send — while she worked on something else.

"The amount of things I can do in a day has now quadrupled. I have other 'versions' of me running in the background doing tasks while I do the things that actually require me."

— Jenna Gaidusek

For designers who are not ready for a fully integrated browser setup, the agent inside ChatGPT offers a more contained version of the same capability — you log in to specific things session by session rather than having persistent access. Less powerful, but also lower risk for those who want to start slowly.

On Security, Privacy, and Deciding What Is Worth It

Jenna is direct about this: giving an AI browser access to your email, calendar, and trade accounts means it can see those things. That is how agentic AI works. She is also direct that she considers this a manageable risk — not because the risk does not exist, but because she has weighed it and decided the time savings outweigh the exposure for her situation.

Her actual approach: she does not put anything sensitive in email that she would not send in a text. She does not store credit cards in the browser. She uses Google's ecosystem specifically because she considers Google's privacy and security controls to be among the strongest available. And she acknowledges that accounts can be hacked regardless of the tool — her husband's Etsy account was compromised through Etsy, not through any AI.

Her advice: read the terms. If you do not understand them, paste them into an AI and ask it to break down your rights and security protections in plain language. Then make your own decision. "Do what you're going to do" — but make it an informed choice, not a passive one.

What AI Actually Replaces — And What It Does Not

Jenna is clear, episode after episode, that the framing of "AI replacing designers" is wrong. This episode is a good illustration of why: every example she gives is a task that was keeping her from design, not a design task itself. Checking emails. Finding links. Adding calendar events. Matching paint colors. Stitching together tools that should have been one workflow.

What she does not want AI to replace: the connection with the client, the judgment about what actually works in a space, the experience of walking into a room and knowing what it needs, the ability to look at a rendered image and immediately see twenty things that are wrong with it.

"I am the overlord and they are doing the minion tasks. My job is to figure out how somebody can use this to better their business, better their life. That's literally all I'm thinking every day when I sit at my desk."

— Jenna Gaidusek

The designers she worries about are the ones spinning their wheels on two-hour tasks that should take five minutes — not because AI threatens their career, but because they are wasting the time they could be spending on the parts of design they actually love. That is the gap this episode is trying to close.

Frequently Asked Questions

Nano Banana is the colloquial name for Google Gemini's image editing tools, which are now embedded directly in the Android camera interface on many devices. For designers, the key capability is product placement: take a photo of an actual product (a rug, a chair, a pendant), provide measurements and a room photo, and the tool places that product into the space at near-accurate scale and material representation. Jenna's current workflow uses it in conversational segments — one or two changes per prompt — to get the most reliable results.

No — and Jenna considers this a feature, not a limitation. AI imagery can look spatially convincing without the furniture actually fitting the real dimensions of the space. A room that looks balanced in an AI mockup may have a sofa that would physically block a door or a kitchen island that violates code clearances. Professional floor plan tools like MyDoma Visualizer still serve as the accuracy foundation. The AI visual communicates the concept and gets client buy-in; the floor plan confirms the concept is actually buildable.

Regular chatbot use is a question-and-answer exchange: you ask, it responds, you do something with the answer. Agentic AI takes action. You tell it to check your email for anything that came in overnight and summarize what needs a response — and it goes to your email, reads it, and gives you the summary. You tell it to add a calendar event, find a conflict, and send a draft invite — and it does all three. The Perplexity Comet browser is Jenna's current primary agentic tool because it has persistent access to her logged-in accounts, making multi-step tasks faster than the session-based agent in ChatGPT.

The Perplexity Comet browser is an AI-native web browser that replaced both Chrome and Edge in Jenna's daily workflow. It combines Perplexity's multi-model search capabilities with agentic functionality — meaning it can access your logged-in accounts (email, calendar, trade websites) and take actions on your behalf, not just answer questions. It includes a persistent assistant panel accessible from any page, voice conversation mode, and the ability to watch the AI complete tasks in real time. At the time of recording, it was invite-only for early access users; it has since expanded more broadly to Perplexity subscribers.

The honest answer is that agentic tools access what you give them access to. If you log into your email through an AI-enabled browser, the AI can see your email. Jenna's framework: do not store sensitive information in email that you would not send in a text. Use platforms with strong privacy controls (Google's ecosystem is her preference). Read the terms of service — and if they are not clear, paste them into an AI and ask it to explain your rights in plain language. Then decide whether the time savings are worth the tradeoff for your situation. There is no universally right answer.

Keep Learning

The DAIly — On-Demand AI Education for Designers

$29.99 a month. Over 150 videos, prompt packs, and beta access to the apps Jenna is building — including the integrated design workflow app described in this episode.

Join The DAIly Certificate Program ↗

Disclaimer: This blog was written using AI as a recap from the recording then edited by the author for accuracy and details.