Bonus: Studio-Quality Headshots (and some other fun things) using ChatGPT, Nano Banana and Veo3
ChatGPT + Google's Gemini Flash 2.5 is like a cheat code
I’m coming to you with a bonus drop this week. The main episode ran a little advanced and niche, so I wanted to balance it with something universal and more accessible to everyone. This is an AI workflow you can spin up today that everyone will find useful.
The inspiration came from a friend who posted a sharp new headshot. So, natually, I thought why not use AI to make my own? That turned into this hands-on walkthrough using ChatGPT and Google’s Gemini Flash 2.5 (Nano Banana + Veo3) to turn any photo into a polished LinkedIn-ready headshot, and then push the same workflow into playful use cases.
Here’s what we cover:
Use ChatGPT to describe a style reference photo and refine it into a reusable prompt.
Feed the prompt + your own photo into Gemini Flash 2.5 to get studio-quality headshots.
Swap backgrounds in seconds (like dropping Pike’s Peak behind me).
Extend the workflow to eBooks, ads, and 3D product mockups.
Have some fun: AI product shots, brand energy drinks, and even animated bar scenes with Wayne Gretzky.
Of course, headshots are just one small use case for this setup. Here are 10 more ways you can use ChatGPT + Gemini Flash 2.5 (Nano Banana + Veo 3):
Dynamic ad creative testing (same subject, varied backdrops, people, etc).
Social profile banners sized for every channel.
On-brand sales decks with photoreal mockups.
Event promo graphics with consistent speaker images.
Ebook cover refreshes and 3D renders.
Personalized thank-you notes turned into custom images.
Company merch mockups (shirts, mugs, hoodies).
Stylized podcast thumbnails that stay on-brand.
Quick explainer clips: static slides turned into short videos.
Internal culture use cases, like fun team images for Slack or all-hands.
Here’s a walkthrough of exactly what we covered
Step 1
Pick a reference headshot you like.
I chose this one of my buddy, Jorge Soto.
Step 2
In ChatGPT, ask for a detailed description of the photo (camera, lighting, background, pose, wardrobe).
Here’s the prompt that I used:
Please describe this headshot as best you can so that I can recreate it using Google’s NanoBanana imaging model. Give me as much detail as you can. Then I will give NanoBanana my own headshot and try to recreate it. You don’t need to mention anything about the glasses or facial hair or anything on the face of the person. I will upload my own headshot to transform into this style of headshot.Step 3
Copy that description.
Go to gemini.google.com and open Flash 2.5, then select “image”
Step 4
Paste your description in.
Adjust variables like expression, pose, or background blur.
Then tell Gemini that you want a new headshot in the style you described.
Finally, upload your own headshot.
Step 5
Hit enter and see what you get!
If you don’t get exactly what you want, re-prompt Gemini.
You can change the background, your clothes, your pose… anything you can imagine.
BONUS:
Bonus fun: Add a second person photo and prompt a bar scene together, then animate the cheers with Veo 3. Layer in voice or sync in ElevenLabs/Latent Sync.
If you found this helpful, be a hero and share!
Hope you enjoyed this mini bonus newsletter!
See you again in a few days for another marketing AI stack.
– Brandon