Professional Photo Prompt: 25 Ready AI Templates

Typing "professional photo" into an image generator is the fastest way to get a generic portrait back: blue suit, blurred office background, stock-photo smile, and plastic-looking skin. The problem is almost never the AI model — it's the prompt. A professional photo prompt that works describes six or seven different layers of the image, and every layer you leave blank gets filled in by the AI with the most likely cliché from its training.

This guide walks through the full structure of a corporate portrait prompt, delivers 25 ready templates organized by goal and profession, explains how to use your own face instead of a stranger's, and closes with the technical finishing touches that separate an image that passes as a real photo from one that screams "AI-generated".

Why most professional photo prompts fail

Image models don't understand intent, they understand description. When you write "35-year-old man, professional photo for LinkedIn", you're giving two pieces of information and omitting about twenty. Camera distance, eye-line height, light direction, light hardness, background color and distance, lens type, skin texture, expression, shoulder angle, final aspect ratio — all of that is left to the model, which will pick the statistical average of whatever it saw labeled as "professional headshot".

That's why results from short prompts all look alike, and all slightly off. The statistical average of corporate photos on the internet includes a lot of stock imagery, a lot of heavy artificial lighting, and a lot of aggressive retouching. Without instructions to the contrary, that's exactly what you get.

There's a second, subtler problem: an overload of empty adjectives. Words like "amazing", "stunning", "ultra professional", "8k", "hyper-realistic", and "award-winning" take up space in the prompt without visually constraining anything. They push the result toward exaggeration — high contrast, high saturation, glossy skin — which is exactly the look that gives away automatic generation. A good prompt is technical and restrained, more like a work order for a photographer than a social media caption.

The third failure is ignoring the context of use. A photo that will become a circular 400 × 400 pixel avatar on LinkedIn needs tighter framing and a cleaner background than a photo that will fill half a page on the company's website. If the prompt doesn't know the image's destination, it gets the framing wrong.

💡 Tip: before writing any prompt, decide where the photo will be used. A small circular avatar calls for a tight shot from the chest up. A banner or "About" page can handle a medium shot with negative space beside it for text.

The anatomy of a professional photo prompt: 7 blocks

Every portrait prompt that delivers a consistent result can be broken down into seven blocks. You don't need to write all of them on every attempt, but the more blocks you fill in, the less the AI improvises. Order matters too: diffusion models give more weight to the start of the prompt, so subject and framing come first.

1. Subject and identity

Describe the person with enough precision to remove ambiguity, without turning the prompt into a police report. Age range, gender (if relevant), skin tone, hair type and length, facial hair, glasses, and distinctive features are enough. Avoid the names of real people and celebrities: besides being ethically problematic, many services block or distort that kind of request.

Examples of useful descriptions: "40-year-old woman, shoulder-length wavy brown hair, light brown skin, thin-framed glasses"; "28-year-old man, dark skin, short curly hair, well-trimmed short beard".

2. Framing, pose, and expression

This is the most-forgotten block and the one that changes the result the most. Vocabulary that works well:

Shot: headshot (head and shoulders), chest-up shot, medium shot (waist up), full body.
Angle: eye level (standard for corporate portraits), slightly above eye level (softens the chin), slightly below (conveys authority).
Rotation: facing forward, three-quarter view, partial profile. Three-quarter with shoulders slightly turned is the most versatile corporate pose.
Gaze: direct at the camera (confidence, ideal for LinkedIn) or slightly off-frame (used in feature and column photos).
Expression: a light closed-mouth smile, a natural open smile with the eyes engaged, a neutral and attentive expression. Ask for "natural smile, eyes slightly crinkled" to avoid the stiff mannequin smile.

3. Clothing

Specify garment, color, and level of formality. "Navy blazer over a white shirt, no tie" is an instruction. "Professional clothing" is not. Some pairings that work well by industry: dark blazer and light shirt for law and finance; plain dress shirt with no blazer for tech; turtleneck sweater or linen shirt for creative fields; white coat over a polo shirt for healthcare. Avoid small patterns and thin stripes — they create artifacts and moiré in most models.

4. Lighting

Lighting is what makes an image look like a photograph instead of a render. Terms models recognize well:

Soft side window light: the safest choice for a natural, pleasant portrait.
45-degree softbox with fill reflector: the corporate studio standard — predictable and clean.
Subtle rim light: separates the hair from a dark background and avoids a flat, cut-out outline.
Rembrandt lighting: a triangle of light on the opposite cheek, more dramatic, good for speakers and executives.
Diffuse overcast daylight: great for an outdoor portrait with a spontaneous feel.

Explicitly ask for "soft, realistic shadows" and "no harsh frontal flash". Direct flash is one of the most common signatures of a generated image and instantly dates the photo.

5. Background

Choose between three families and say which: a solid studio backdrop (medium gray, charcoal gray, off-white, petrol blue), a blurred environment (glass-walled office, bright hallway, brick wall, urban greenery), or open outdoors. Either way, define the separation: "background two meters behind the subject, slightly blurred" avoids the pasted-on look where the person seems stuck to the wall.

For a social media avatar, a solid or near-solid background wins almost every time — it survives the circular crop and a 48-pixel thumbnail.

6. Camera, lens, and depth

You don't need to mention a camera brand, but naming a lens changes the geometry of the face. Corporate portraits usually use the equivalent of an 85mm or 105mm lens with an aperture between f/2 and f/4. Wide lenses (24mm, 35mm) distort the nose and ears when shot up close — only use them for a deliberate editorial effect. Adding "shallow depth of field, but the whole face in focus" avoids the common mistake of one eye coming out blurry.

7. Finish, color, and aspect ratio

Close the prompt with the treatment: "natural color, true skin tones, preserved skin texture with visible pores, no excessive smoothing, moderate contrast". And state the aspect ratio: 1:1 for an avatar, 4:5 for a vertical post, 3:2 or 16:9 for a banner or website.

💡 Tip: "preserved skin texture, visible pores, no skin retouching" is probably the highest-value instruction in the entire prompt. Skin that's too smooth is what the human eye identifies as fake the fastest.

The ready-made formula: a template you can copy and adapt

Putting the seven blocks together, the skeleton looks like this:

Professional portrait of [AGE + DESCRIPTION OF THE PERSON], [SHOT AND ANGLE], [POSE], [EXPRESSION], wearing [CLOTHING WITH COLOR], lit by [LIGHTING SETUP], [BACKGROUND AND DISTANCE], shot with a [XXmm] lens at [f/X], shallow depth of field with the whole face in focus, natural color and true skin tones, preserved skin texture with visible pores, no excessive smoothing, aspect ratio [X:Y]

And filled in, for the most common case — a LinkedIn avatar:

Professional portrait of a 38-year-old woman, light brown skin, shoulder-length wavy brown hair, chest-up framing at eye level, shoulders turned three-quarters with the face toward the camera, a light natural smile with the eyes engaged, wearing a charcoal-gray blazer over a white shirt, lit by soft light from a large window on the left with a fill reflector on the right and a subtle rim light on the hair, uniform medium-gray background two meters away and slightly blurred, 85mm lens at f/2.8, shallow depth of field with the whole face in focus, natural color and true skin tones, preserved skin texture with visible pores, no excessive smoothing, 1:1 aspect ratio

It looks long, and it is — but every part of it takes one more decision out of the AI's hands. Once you've built yours, save the prompt in a text file and change only one block at a time between attempts. Changing five things at once makes it impossible to know what actually improved the result.

25 ready-made professional photo prompts

All the prompts below are written in plain English, which works well with the latest multimodal models. On older or purely diffusion-based tools, minor rewording sometimes renders one or two extra points of fidelity. Swap in your own description of the person and keep the rest.

Classic corporate for LinkedIn

Corporate portrait chest-up, three-quarter angle, light smile, navy blazer over a white shirt with no tie, 45-degree softbox light with soft fill, uniform medium-gray background, 85mm f/2.8, preserved skin texture, 1:1 aspect ratio.
Professional portrait in a modern office, chest-up shot, direct gaze at the camera, confident and calm expression, light-blue dress shirt with buttoned sleeves, natural light from a wide window on the right, blurred glass-and-wood background three meters back, 105mm f/2.5, 1:1 aspect ratio.
Executive portrait on a dark background, Rembrandt lighting with rim light on the shoulder, charcoal suit and sober tie, neutral and attentive expression, tight chest-up shot, 85mm f/2, moderate contrast, no skin smoothing, 1:1 aspect ratio.
Relaxed professional portrait, medium shot from the waist up, person with arms loosely crossed leaning against a light-colored wall, thin gray sweater, diffuse overcast daylight coming in from the side, off-white background, 50mm f/2.8, natural color, 4:5 aspect ratio.
Corporate portrait seated at a desk, body turned three-quarters with the face toward the camera, white shirt with a beige blazer, relaxed hands on the desk, soft side window light, heavily blurred bright office background, 85mm f/2.8, 3:2 aspect ratio.

Resumes and job applications

Sober resume photo, head-and-shoulders headshot, smooth white background with no cast shadow, large diffuse frontal softbox with bottom fill, navy-blue dress shirt, neutral and approachable expression, 105mm f/4, whole face in focus, 1:1 aspect ratio.
Portrait for an administrative-role resume, chest-up shot, uniform light-gray background, black blazer over a light-colored blouse, subtle closed-mouth smile, clean two-source studio lighting, no strong forehead reflection, 85mm f/4, 3:4 aspect ratio.
Portrait for a first professional experience, young 22-year-old person, subtle plaid shirt with no blazer, light-colored wall background with slight texture, soft window light from the left, open and friendly expression, 50mm f/2.8, 1:1 aspect ratio.
Formal portrait for a civil-service exam or credential photo, pure white background, uniform lighting with no shadows on either side of the face, plain dress shirt, neutral expression, direct gaze, head centered with even space above, 105mm f/5.6, 3:4 aspect ratio.

Authority: speaker, executive, consultant

Editorial portrait of a speaker, medium shot, body leaning slightly toward the camera, dark open-collar shirt, dramatic side lighting with controlled shadow, deep black background, rim light defining the shoulder outline, 85mm f/2, 4:5 aspect ratio.
Portrait of a consultant in a workplace setting, chest-up shot, person standing with arms crossed, petrol-blue blazer, blurred meeting-room background with late-afternoon light coming in, 85mm f/2.2, slightly warm and natural color, 3:2 aspect ratio.
Portrait of a senior executive, tight shot, direct and steady gaze, neutral expression with slight tension at the corners of the eyes, structured white blazer, high frontal beauty-dish lighting with fill, light-gray gradient background, 105mm f/4, 1:1 aspect ratio.
Author portrait for a book jacket, chest-up shot in black and white, gaze slightly off-frame, plain shirt, hard window light cutting across half the face, dark background, fine film grain, 85mm f/2, 4:5 aspect ratio.

Tech and creative fields

Portrait of a software developer, chest-up shot, plain dark t-shirt with no print, open-plan office background with blurred lights behind, soft side light with a slight blue tint in the background, relaxed expression, 85mm f/1.8, 1:1 aspect ratio.
Portrait of a designer in a studio, medium shot, gray turtleneck sweater, exposed concrete-wall background two meters back, large window light on the left with a visible shadow on the right, desaturated and natural color, 50mm f/2, 4:5 aspect ratio.
Portrait of a photographer or videomaker, chest-up shot outdoors at golden hour, open linen shirt over a t-shirt, sun as rim light from behind with soft fill in front, blurred urban background, 85mm f/1.8, 3:2 aspect ratio.
Portrait of a marketing professional for an "About" page, medium shot with negative space on the right for text, vivid solid-color blouse, uniform bright background, soft frontal light, smiling and open expression, 70mm f/4, 16:9 aspect ratio.

Healthcare, law, and finance

Portrait of a female doctor, chest-up shot, white coat over a light-blue blouse, discreet stethoscope in the pocket, heavily blurred bright clinic background, clean and even lighting, warm and calm expression, 85mm f/2.8, 1:1 aspect ratio.
Portrait of a dentist or physical therapist, standing medium shot, light-colored uniform, white clinic background with out-of-focus equipment, neutral diffuse light with no hard shadows, natural smile, 85mm f/2.8, 4:5 aspect ratio.
Portrait of a lawyer, chest-up shot, dark suit and white shirt, blurred legal-bookshelf background three meters back, soft and slightly warm side light, serious and trustworthy expression, 105mm f/2.8, 1:1 aspect ratio.
Portrait of a finance professional, tight shot, navy blazer, solid blue-gray background, two-point studio lighting with moderate contrast, direct gaze, no broad smile, preserved skin texture, 105mm f/4, 1:1 aspect ratio.

Personal brand and social media

Portrait of a content creator, chest-up shot, plain solid-color t-shirt, uniform vividly painted wall background, soft and clean frontal light, open and natural smile, slightly saturated color, 50mm f/2.8, 1:1 aspect ratio.
Portrait for a social media banner, medium shot on the left side of the frame with empty space on the right, neutral clothing, dark gradient background, rim light on both sides, 85mm f/2.8, 16:9 aspect ratio.
A sequence of three portraits of the same person, same outfit and same light, varying only the expression between neutral, a light smile, and an open smile, uniform gray background, 85mm f/2.8, 1:1 aspect ratio in each.
Spontaneous outdoor professional portrait, chest-up shot walking along a tree-lined sidewalk, light blazer, diffuse late-afternoon light filtering through the leaves, heavily blurred urban background, subtle hair movement, 85mm f/1.8, 4:5 aspect ratio.

💡 Tip: always generate a batch of four to six variations of the same prompt and choose afterward. Variation between images from the same prompt is large, and the difference between the first and fifth attempt is often bigger than the difference between two competing models.

How to use your own face — not a stranger's

The prompts above generate a plausible person, not you. For a profile photo, resume, or company website, that's a serious problem: presenting a face that isn't yours in a professional context is misleading and usually backfires at the first in-person meeting or video call. There are three honest paths.

Path 1: start from a real photo of yourself

Current multimodal tools accept a reference image along with the text. Instead of generating from scratch, you upload a selfie and write an editing instruction, keeping your identity and changing only what needs to change. Instructions that work well: "keep exactly the face, hair, and features from this photo; replace the background with a uniform medium-gray studio backdrop; swap the t-shirt for a light-blue dress shirt; redo the lighting as a soft 45-degree softbox; preserve the natural skin texture".

This is the fastest and most faithful path. The better the starting photo — sharp, well-lit, looking at the camera, no sunglasses or hats — the better the result.

Path 2: train a model on your own photos

Dedicated headshot services ask for 8 to 20 photos of you and train a small model that learns your face, delivering dozens of variations in clothing, background, and lighting. It's the option with the best consistency when you need many photos, or when a company wants to standardize the whole team's photos. For the upload, choose photos with varied angles, different expressions, different outfits, and different backgrounds — repeating ten selfies from the same day teaches the model to reproduce that day, not your face.

Path 3: take the photo and use AI only for the finishing touches

Often the best professional photo is a real phone photo, corrected afterward. A modern phone's front camera, window light, a plain wall as a backdrop, a tripod or improvised support, and a timer solve 80% of the work. After that, what's left is technical: cropping, sharpening, cleaning up the background, and exporting at the right size — none of which requires generative AI. If you have a good photo that's out of focus, the guide on how to sharpen a blurry photo covers the most common case.

Where to run the prompt: how each tool responds

The same prompt produces quite different results depending on where you run it. The table below summarizes the typical behavior of each family of tools for professional portraits, without getting into pricing — which changes constantly.

Tool	How it responds to the prompt	Best for
Gemini (Nano Banana)	High text adherence and good face preservation when you send a reference photo	Editing your own photo while keeping your identity
ChatGPT (GPT Image)	Understands long natural-language prompts and accepts conversational corrections	Iterating through dialogue, adjusting one block at a time
Midjourney	Very strong aesthetic, but reinterprets instructions and leans cinematic	Editorial-style authority portraits for speakers
Flux	High photographic realism and good response to lens and lighting terms	A headshot that needs to look like a real photograph
Leonardo / Firefly	Panel controls (aspect ratio, reference strength) in addition to text	People who prefer sliders over pure prompting
Headshot services	The prompt is hidden; you pick from ready-made styles after uploading photos	Standardizing photos for an entire team

If you want a broader overview of platforms before choosing, the comparison of the best AI tools to create images covers the strengths and limitations of each.

Negative prompts: what to ask to keep out

Half the quality of a generated portrait comes from what you rule out. On tools with a negative-prompt field, use a short, specific list; on tools without that field, write the exclusions as a sentence at the end of the main prompt.

Skin and retouching: plastic-looking skin, excessive smoothing, oily shine on the forehead, airbrushing, artificial catchlights in the eyes.
Anatomy: deformed hands, extra fingers, uneven teeth, asymmetrical ears, elongated neck.
Optics: wide-lens distortion, one eye out of focus, strong chromatic aberration, heavy dark vignette.
Aesthetics: exaggerated saturation, aggressive HDR, extreme contrast, social-media filter, watermark, text on the image, frame/border.
Composition: top of the head cropped off, face touching the edge, background stuck to the subject, tiny busy patterns on clothing.

A more complete and reusable negative list, with an explanation for each term, is in the guide to negative AI prompts to improve images.

9 signs that give away an AI-made professional photo

Before publishing, open the image at full size and look for these points. They account for the vast majority of images that "look off" without the viewer being able to say why.

Poreless skin. Absolute uniformity of texture, with no tonal variation on the cheeks or nose.
Inconsistent eye reflections. The catchlight in each eye points in a different direction, or only appears in one eye.
Fused teeth. A row with no clear separation between teeth, or the wrong number of them.
Ears and glasses. Glasses arms that cut through the ear, disappear into the hair, or have a different thickness on each side.
Melting hair. Strands that fade into the background instead of ending, or a suspicious outline around the head.
Impossible clothing. A collar that doesn't close, a button with no matching hole, a lapel that switches sides, a seam that breaks off midway.
Jewelry and accessories. Mismatched earrings, a ring that changes finger, a badge with illegible text.
Warped background geometry. Door and window lines that aren't parallel, garbled text on signs, plants with fused-together leaves.
Light with no source. The shadow on the face comes from the left, but the glow on the wall comes from the right.

If you want to understand these patterns in more depth — including for evaluating other people's images — it's worth reading how to know if an image was made by AI.

The finishing touches: what to do after generating the image

No image comes out of the generator ready to publish. This is the sequence that turns a raw file into a profile photo that works on any platform.

Crop to the right ratio. An avatar wants 1:1 with the head slightly above center and even margins on the sides. Use the crop image tool and then check whether the face survives the circular crop that social networks apply.
Fix the background if needed. If the background came out cluttered or with a stray object, the remove background tool isolates the person and lets you swap in a solid color — the safest background for a small thumbnail.
Adjust sharpness carefully. A slight sharpness boost brings back detail in hair and eyes; overdoing it creates a halo. The sharpen image tool fixes this in a few clicks.
Resize for the destination. LinkedIn and most networks work well with 800 × 800 pixels as the source. The resize image tool scales it down without jagged edges.
Compress before uploading. A profile photo doesn't need 4 MB. The image compressor drops the file size while keeping visible quality.
Clean the metadata. If the image ever passed through your phone, it may carry location and device model data. The remove EXIF metadata tool clears that before publishing.

For fine-tuning brightness, contrast, and cropping in the same place, the online image editor covers the whole workflow with nothing to install. And if your specific goal is LinkedIn, the guide on LinkedIn photos covers the recommended dimensions and framing choices for the platform.

Finish your professional photo right in your browser

Crop, adjust, resize, and export your profile photo for free — no software to install and no image ever sent to a server.

Open the image editor

Final checklist before publishing

Go through these nine items. If they all pass, the photo is ready.

The face is recognizably yours, compared with a recent real photo.
The clothing matches what you'd actually wear in a real meeting in your field.
The framing survives the circular crop and a small thumbnail.
The skin has visible texture at 100% zoom.
Eyes, teeth, ears, glasses, and jewelry have passed the nine-sign inspection.
The light has a single direction, and the background agrees with it.
The background is clean, with no legible text or distracting object.
The file is 1:1, weighing in under a few hundred kilobytes.
The metadata has been removed.

One final note on transparency: using AI to improve the lighting, background, and finish of a real photo of yourself is editing, and that's accepted in any professional context. Generating a face that isn't yours, substantially changing your appearance, or creating a scene that suggests an experience you never had is something else entirely — and the cost of being found out is much higher than the benefit of a nicer-looking photo. Keep the photo recognizable and you won't run into any trouble.

Frequently asked questions

What's the best prompt for a professional photo?

There's no single prompt, there's a structure. The best prompt describes seven blocks in order: subject, framing and pose, clothing, lighting, background, lens and aperture, and finishing with aspect ratio. A short prompt like "professional photo for LinkedIn" leaves twenty decisions in the AI's hands and returns the average internet cliché. Copy this guide's template, fill the seven blocks with your own details, and change one block at a time between attempts.

Do I need to write the prompt in English?

Not necessarily. Recent multimodal tools understand other languages quite well, including technical photography terms. On older diffusion models or interfaces built on open models, English still tends to render slightly more faithfully, especially for lighting vocabulary. A good strategy is to write in your own language, evaluate the result, and if prompt adherence is low, translate and compare.

Can I use an AI-generated professional photo on LinkedIn or my resume?

Yes, as long as the person in the image is you and remains recognizable. The recommended path is to send a real photo of yourself as a reference and ask the AI to adjust only the background, clothing, and lighting, keeping face, hair, and features intact. Generating a made-up face to use as a professional profile photo is misleading and falls apart at the first video call or in-person interview.

How many photos do I need to send for the AI to learn my face?

Personal model training services usually ask for between 8 and 20 images. What matters most isn't quantity but variety: different angles, different expressions, different outfits, different light and backgrounds, all with the face sharp and unobstructed. Ten selfies taken in the same minute teach the model to reproduce that specific moment and produce repetitive results. If you only want one or two photos, editing a real photo of yourself is enough and much faster.

Why does the skin end up looking plastic?

Because most models were trained on a lot of retouched imagery, and adjectives like "hyper-realistic", "8k", and "stunning" push the result in that direction. The fix has two parts: including instructions in the main prompt like "preserved skin texture, visible pores, no excessive smoothing, moderate contrast", and listing retouching terms in the negative prompt — plastic-looking skin, airbrushing, oily shine, skin smoothing. It also helps to ask for soft window light instead of frontal flash.

What size and format should I use for the final photo?

For a profile photo, generate or crop to a 1:1 ratio and export at around 800 × 800 pixels — enough resolution for high-density screens and light enough to load fast. JPG is the standard choice for portraits, since it compresses skin-tone gradients well; PNG only makes sense if you need a transparent background. Before uploading, compress the file and remove the EXIF metadata.

A well-written professional photo prompt is, at its core, a photography brief. You're not asking the AI to have good taste — you're telling it exactly where to put the light, how far back to set the background, which lens to use, and what not to do with the skin. Start with the template, change one block at a time, generate in batches, and save the last five minutes for the finishing touches. It's that last, tedious, technical step that makes the difference between an interesting image and a profile photo that passes for real.