Co-Founder, The House of Curiosity | Curator, Best of AI | CPP: ImagineArt, Flora, Caimera, Higgsfield | Making Complex Tools Useful

I'm Israel Ayliffe, an AI explorer obsessed with finding out what these tools can really do.

I spend my time experimenting with ChatGPT, Claude, MidJourney, and whatever else I can get my hands on... then turning what I learn into visual stories, prompt frameworks, and practical tools that help others use AI without feeling overwhelmed.

I don't come from a technical background. I just love figuring things out and showing others what’s possible when you mix curiosity with the right tools.

If you're looking to make AI feel more accessible, useful, or creatively exciting, I'm always up for a good conversation.

Ever liked a prompt so much, you cloned it?

Here's exactly how.

I wrote a cinematic motion-blur prompt for Nano Banana 2. Black-and-white, Leica SL2, slow shutter panning, model standing still while the crowd blurs around her. It worked.

So I pulled the prompt apart to find the pattern underneath.

The structure is held together by six pillars: camera rig, slow shutter panning, absolute model stillness against crowd blur, layered coat-over-dress wardrobe, flat overcast lighting, and realism anchors paired with safety rails.

Everything else is a variable slot: angle, location, model, pose, wardrobe combination, blur metaphor.

Here's the template:

[STYLE DECLARATION: Cinematic black-and-white / Luxury monochrome] [FORMAT: high-fashion campaign image / campaign portrait / editorial photograph], [CAMERA: shot on Leica SL2 with a 50mm Summilux at f/2.8], [TECHNIQUE: slow shutter panning] from [ANGLE: tight chest-up / high frontal / behind model / side profile at hip height / extreme low / slight overhead / through blurred foreground figures]. [MODEL: A {build/ethnicity descriptor} female model] [STILLNESS: stands absolutely/perfectly/completely still] [LOCATION: in/on/under/beside {urban architectural setting}], [BODY LANGUAGE: specific stance, limb position, head tilt, gaze direction], wearing [LAYER 1: a {texture/volume descriptor} long coat with {defining detail}] over [LAYER 2: a {fit descriptor} dress/gown] and [FOOTWEAR: {boot style}]. [POSE DETAIL: one specific gesture or hand placement that adds narrative]. [CROWD MOTION: The crowd {verb} around her as/like {poetic blur metaphor}]. [LIGHTING: Flat overcast daylight], [MOOD: {two-word gothic/spiritual/luxurious tone}], [FILM TEXTURE: fine grain/fine film grain], [REALISM ANCHORS: sharp eyes, visible pores/real skin pores, realistic skin and textile rendering], [SAFETY RAILS: no face distortion/no warped limbs/no distortions/no warped anatomy], [ASPECT RATIO: 4:5].

Feed this template into ChatGPT or any LLM and ask for variations across every slot. Different angles, different cities, different wardrobe architectures, different crowd metaphors. Same skeleton, different skin.

Run every variation in Nano Banana 2. Below are my results.

My take: getting the prompt right is only the start. The real move is chaining tools. One LLM to vary the template, one image gen tool to render the output. That's how a single idea becomes a series.

#Nanobanana2 #AIImageGeneration #PromptEngineering #AIWorkflow #AIFashion #BlackAndWhitePhotography

Israel Ayliffe

I’m Israel Ayliffe, a Certified Nursing Assistant with a deep commitment to compassionate caregiving and an enthusiasm for applying artificial intelligence in practical, non-technical ways. Inspired by personal experiences, I’ve dedicated my career to healthcare while also exploring how AI tools like ChatGPT, Claude, and Midjourney can empower caregivers and streamline daily tasks.

My journey is about learning and sharing. Each day, I experiment with AI, creating simple prompt templates and practical tips that help make these tools accessible to those who aren’t tech experts. I believe in the power of AI to make a real difference, whether in patient care, stress relief, or even creative projects.

Let’s connect if you’re interested in how AI can enhance caregiving, improve workflows, or simply make life a little easier.

I uploaded a selfie to an AI headshot prompt I built.

Not a headshot app. A prompt with 4 stages that run before a single prompt gets generated.

Stage 1: Image Analysis
Describe what's in the photo first. Angle, lighting, clothing, background, problems. If it doesn't know the angle is wrong, it can't fix it.

Stage 2: Occasion Inference
Read the clothing to figure out what kind of headshot fits. Denim jacket reads creative, not corporate. No back and forth. Just inference.

Stage 3: Clothing Decision
Act on that inference. Does the clothing work for the occasion or should it swap? Denim for a creative bio page? Keep. Gym tank top for a corporate headshot? Change.

Stage 4: Headshot Variations
Different backgrounds, lighting setups, and framing. All headshots. No experiments.

Here's a mini version of the prompt you can use right now:

"Analyze the uploaded photo in detail: subject, hair, clothing, angle, lighting, background, and problems to fix. Then infer what kind of headshot fits based on the clothing (business formal = corporate, smart casual = creative/startup, very casual = social media). Decide whether to keep or change the clothing based on the occasion. Then generate 5 headshot editing prompts that each vary in background, lighting, or framing. Every prompt must start with 'Using the uploaded [explicit description of the photo]...' correct the angle to eye-level, replace the background, apply studio lighting, specify 8K detail with natural skin texture, and end with 'Do not change aspect ratio.'"

Our team at The House of Curiosity needed headshots for something we're putting together. I didn't have a photographer handy. This got me to about 90% likeness. Not perfect. But usable.

Full detailed template in the comments.

What stage would you add?

#AIHeadshots #PromptEngineering #AIImageGeneration #TheHouseOfCuriosity #GeminiImage #AIWorkflow

Experiment: Illustration to photorealistic photo.

Tool: Nano Banana Pro (Gemini 3 Pro Image, image editing mode).

The transforms aren't always clean. The model shifts ethnicities, adds artifacts, drifts faces entirely.

Most of that is fixable with better prompting. What prompting can't fix, the next model will.

Recommended combo: Niji 7 or Recraft 4 for the illustration. Nano Banana Pro for the photo transform.

Full prompt template below. Copy it, fill the brackets, run it.

---

ILLUSTRATION TO PHOTO TEMPLATE

Using the uploaded illustration of [DESCRIBE: subjects, props, clothing, colors, background, style], transform this illustration into a photorealistic photograph.

Preserve exact identity and design DNA: [3-5 ANCHORS: face shape, hair, signature accessories, outfit details, palette, pose].

Preserve composition and spatial arrangement: [framing, camera distance, subject placement, negative space].

Replace illustrated rendering with photographic reality. Real materials, natural surface texture, physically accurate light. Skin with visible pores. Fabrics with weave and drape. Hair with individual strands and flyaways.

Match the illustration's lighting intent: [WARM/COOL], [DIRECTIONAL/AMBIENT], [TIME OF DAY]. Same shadow direction and mood. Realistic depth of field.

Preserve all elements. Add nothing new. This should look like a real photo of the same scene, not a filter. Do not change aspect ratio.

---

#NanoBananaPro #MidJourney #GeminiImage #AIArt #PromptEngineering #ImageEditing #AIImageGeneration

Two Inks

Whole worlds in two inks

Prompt: Illustration, screenprint portrait, a single man with wavy hair and a simple shirt collar, no hands visible; emotion: thoughtful; action: gaze downwards, slightly off-camera; details: bold nose shape, minimal features, one geometric dot accent. Influences: pop printmaking + mid-century. Comp: head on upper right third, large negative space left; Palette: orange and deep green on cream paper; Rendering: flat layers with slight misregistration; Surface: uncoated paper with visible fibers. --chaos 6 --ar 4:5 --sref 4086557345 525203247 --profile bp59eui --sw 150 --stylize 230 --weird 2 --niji 7

Prompt: Illustration, stencil printmaking, a single rhinoceros in chunky simplified forms; emotion: solid; action: rhino facing right, head lowered slightly; details: minimal horn shape, no other animals. Influences: folk poster silhouettes. Comp: rhino on lower left third, negative space above; Palette: orange background, deep green rhino, cream paper; Rendering: rough edges, ink speckle; Surface: fibrous paper texture. --chaos 6 --ar 2:3 --sref 4086557345 525203247 --profile bp59eui --sw 170 --stylize 260 --weird 2 --niji 7

Prompt: Illustration, folk screenprint portrait, a single woman in a wide-brim hat, face simplified, no hands visible; emotion: confident; action: gaze straight ahead but slightly off-center; details: hat brim as bold shape, simple earring circle. Influences: pop printmaking + riso editorial. Comp: head placed low in frame with hat filling top negative space; Palette: orange hat/background, deep green shadow blocks, cream paper; Rendering: flat inks with misregistration and speckle; Surface: fibrous matte paper. --chaos 6 --ar 4:5 --sref 4086557345 525203247 --profile bp59eui --sw 160 --stylize 260 --weird 2 --niji 7

#aiart #illustration #midjourney #MidjourneyAI #printmaking #AIImageGeneration

Cinematic motion-blur photography shot on a Leica SL2 with a 50mm Summilux at f/2.8, captured from a chest-level frontal angle using slow shutter panning. Lighting setup: flat overcast daylight acting as soft diffusion, allowing long blur trails without harsh contrast. Color tone black and white inspired by A24 films. Subject: a 28-year-old female model standing completely still, sharp eyes, short curly hair, symmetrical face, real skin pores visible, wearing understated fashion. Crowd action: hundreds of commuters rushing past in all directions, bodies stretched into smooth horizontal and diagonal blur. Emotion: quiet rebellion through stillness. Aesthetic: editorial calm vs. chaos. Post: clean motion physics, fine grain, no glitches, no face distortion, no warped anatomy.  Aspect Ratio 4:5 

#1image #BestofAI #Nanobanana2 #AIImageGeneration

Signal 

Make room for the point 

Exqmple Prompt: Illustration, minimalist ink sketch, a single snail with a spiral shell, cute and ultra-clean; details: antennae as two tiny lines, one ground stroke. Influences: minimalist micro-nature illustration. Comp: snail tiny in lower-left, huge white space; Palette: black ink + pale beige wash. --chaos 6 --ar 3:2 --sref 145151691 --profile bp59eui --sw 140 --stylize 150 --weird 1 --niji 7

#aiart #minimalism #Niji  #AIImageGeneration #Midjourney #MidjourneyAI

Midjourney is running a V8 rating party right now. The focus it seems is text in images.

What I saw is encouraging.

Clean typography. Simple. Readable. Sometimes funny.

Text rendering has been one of MJ's many weak spots for a while. If V8 tightens that up, it changes what you can build with it.

Link to the rating party in the comments.

*Images from the MJ community rating pool, not mine.*

#Midjourney #MidjourneyAI #AIImageGeneration #AIArt #TextInImage #GenerativeAI #AITools

Regal Blur

Stillness Before the Roar

Prompt: Realistic photo of a majestic West African lion on a sun-baked savanna rise above sparse acacia shrubs, viewed from a low angle. The lion is positioned on the left third of the frame with a powerful and intense gaze, nostrils flaring as it gives a low, rumbling growl, wind pushing through its shorter, rugged mane. Golden hour lighting, dramatic and dynamic shadows with studio lighting-style shadows and strong contrast. Clear and detailed linework, volumetric shading, high contrast, sharp and well-defined image, realistic drawing style --chaos 10 --ar 2:3 --seed 2995800647 --raw --sref 8453092561 --profile cdwsdwn 3qesjae 8yt91lu --stylize 250

#1image #aiart #midjourney #MidjourneyAI #wildlifeart #AIImageGeneration

A Japanese movie poster with big bold title "中空スパイラル" in black bold font on white Background the 70s bizzarre horror, bitmap printing style, Black and white Anime style, --chaos 10 --ar 2:3 --seed 2658071408 --raw --sref 3417434064 6634792263 --profile cdwsdwn 2lgasqw 8yt91lu --stylize 200


#1image #MidjourneyAI #AIArt ##aiimagegeneration

taplio

Israel Ayliffe's Linkedin Analytics

Israel Ayliffe

Performance ScoreIsrael Ayliffe's LinkedIn Performance

Israel Ayliffe's Best Posts (last 30 days)

Ready to grow your LinkedIn presence?

taplio

.css-6ytdur{background-image:linear-gradient(284deg, rgba(39, 43, 72, 1) 0%, rgba(33, 47, 183, 1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Israel Ayliffe's Linkedin Analytics

Israel Ayliffe

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Israel Ayliffe's LinkedIn Performance

Israel Ayliffe's Best Posts .css-1emoo84{font-size:var(--chakra-fontSizes-md);font-weight:400;color:var(--chakra-colors-gray-600);}(last 30 days)

Ready to grow your LinkedIn presence?

Israel Ayliffe's Linkedin Analytics

Performance ScoreIsrael Ayliffe's LinkedIn Performance

Israel Ayliffe's Best Posts (last 30 days)