Great initiative! 🔥

Looks like we're going to hell this week, so I'm switching to puppies! 🐶😅

As I misinterpreted the challenge (reimagining the direction first)... I've looked at the image again to try and replicate the original by prompting alone (no image inputs, and no nano-banana editing)! To try and shortcut a version 2, I have a few custom GPTs that I use for reverse engineering images which give different prompt structures, which I test, then tweak and refine adding details until I can get something closer, or that I'm happy with. I tried with midjourney first, but then switched to Seedream 4.0 and here's where I landed (my full prompt / image output below): Ultra-realistic cinematic comedy portrait, perfectly centered main character, medium-close framing in 16:9 format, eye-level camera position, 50mm lens, subject occupying approximately 70% of frame height, shallow depth of field, luxury commercial fashion campaign photography, subtle film grain. A hyper-serious retro athlete from the late 1970s standing directly in front of camera, leaning slightly forward with aggressive competitive energy, mouth fully open in an intense scream, visible upper and lower teeth and tongue, deep forehead tension lines, direct eye contact. Dark brown athletic mullet with center part, smooth texture, tucked behind ears, extending below the neck. Thick vintage mustache, otherwise clean-shaven. Oversized amber-tinted aviator sunglasses with thin gold metal frame. Wide mustard-yellow terry cloth athletic headband. Wearing a luxury cream-white zip-up tennis track jacket with structured collar, premium athletic knit fabric, subtle green racing stripes along the sleeves, small crocodile chest logo, white crew-neck shirt underneath. Everything beyond the main character is soft focus. The environment is a meticulously landscaped retro mini-golf attraction designed with strong graphic geometry and visual asymmetry. Immaculate emerald-green artificial turf occupies the midground. A softly contoured pale sand bunker sits on the left side of the course, partially blurred by depth of field.

Thanks @Brian Terry I can definitely see that interpretation, as that was my first read. After looking again yesterday, I fully understand why the challenge was framed around replication. As a training exercise, it forces us to properly analyse the image: composition, styling, light, colour, perspective, character and prompt structure. In a real client brief I agree — we’d ask more questions, explore the brand world and avoid copying something quite so closely. But as an exercise, I can see the value in observing and reconstructing first, because controlled translation becomes easier after that. I just rushed straight to the art-direction interpretation because that’s where my head naturally goes. Useful reminder to slow down, unbundle the image properly, and not skip straight to the fun bit! 😅

@Jon Sickmeyer love the vibe here anyway! 🥳

I really like what you've done here! 👏