AI Image Creator for Social Media

Posted 4 days ago

Worldwide

Summary

AI Image Generation Specialist — DTC Eyewear Brand (Ongoing, Part-Time) Overview SunHeist is a direct-to-consumer polarized sunglasses brand with a strong visual identity and a fully built AI image generation system. We're looking for a skilled AI image operator to produce lifestyle and active-lifestyle imagery of our frames — running and extending prompt systems we've already developed and proven. This is not a “figure it out from scratch” role. You'll be handed a defined persona, product reference images, and production-tested prompt architecture. Your job is to generate strong candidate imagery, then drive the selected image to a clean, brand-accurate, production-final output. The Opportunity The initial set is the first step in a large and ongoing project. We are building an extensive image library for the SunHeist brand across many frame styles, personas, and moments, and this is the beginning of that effort — not a one-off assignment. The right creator grows with us: helping build the library over time and directly influencing the creative direction of the SunHeist brand. We're looking for a long-term partner, not a single delivery. How the Work Runs We provide two things up front: a defined persona (the model's look and styling) and product reference images of the frame in the relevant colorway. From there, the work runs in two stages: Stage 1 — Build the set. Generate a small set of candidate images of the persona wearing the frame, across both lifestyle moments and active moments. We want range and strong composition here — enough quality options to choose from. Stage 2 — Finalize the select. We select the strongest image. You then fine-tune it to production-final: correcting the product (frame geometry, lens color and gradient), the logo, and remaining fine detail until the output is exact and brand-accurate. The make-or-break of this role is Stage 2. Getting the style right and rendering the logo correctly and cleanly in the final image is critical. AI reliably botches eyewear logos and frame detail — the operators we keep are the ones who can consistently drive an image to a clean, logo-correct, on-brand final. What You'll Do • Generate lifestyle and active-moment imagery of a provided persona wearing our frames (primary tool: Nano Banana Pro; comparable tools fine if you adapt quickly) • Deliver curated candidate sets for selection — strong selects, not raw dumps • Take the selected image to production-final: exact frame geometry, correct lens color/gradient, and a clean, accurate logo • Maintain persona and product consistency across every image in a set • Iterate quickly on feedback (typical revision turnaround: 24–48 hours) • Document what works: prompt refinements, failure patterns, workarounds What We're Looking For • Demonstrated experience generating product-accurate AI imagery — ideally physical consumer products (eyewear, apparel, accessories, CPG) • Proven ability to render logos and fine product detail correctly in a final image — this is the highest-value skill for this role • Fluency with reference-image + prompt workflows (image-to-image, product locking, style and persona consistency) • A portfolio showing consistency across a series — the same subject and product held steady across multiple scenes, not just impressive one-offs • A strong eye for commercial photography: lighting, composition, lens behavior, realistic skin and environments • Reliable communication and predictable turnaround Nice to Have • E-commerce / DTC brand experience • Basic retouching (Photoshop, Illustrator or equivalent) for finishing passes • Experience producing imagery for Meta ads and landing pages Engagement & Compensation • Start with a paid test project (details provided to shortlisted candidates) • Then ongoing part-time engagement, hourly or per-deliverable — open to your preferred model • Estimated volume: [10 images / batch 4-8+ batches per month] How to Apply 1. Share 3–5 examples where you rendered the same product across multiple scenes with a correct, legible logo in the final image 2. In 2–4 sentences, describe your process for taking a selected image to production-final — specifically how you correct product detail and logo 3. Tell us your preferred pricing model and typical turnaround for a 10-image batch

  • Less than 30 hrs/week
    Hourly
  • 1-3 months
    Duration
  • Expert
    Experience Level
  • $30.00

    -

    $70.00

    Hourly
  • Remote Job
  • Ongoing project
    Project Type
Skills and Expertise
Mandatory skills
Midjourney AI
Stable Diffusion
Activity on this job
  • Proposals:20 to 50
  • Last viewed by client:2 days ago
  • Interviewing:
    0
  • Invites sent:
    0
  • Unanswered invites:
    0
About the client
Member since Jul 21, 2026
  • United States
    Winder3:58 AM
  • 2 hires, 2 active

Explore similar jobs on Upwork

Ebook
Copywriting
Brochure
Graphic Design
English
Pool and Hardscapes DesignerHourly‐ Posted 2 months ago
Adobe Photoshop
Graphic Design
Web Design
Adobe Illustrator

How it works

  • Post a job icon
    Create your free profile
    Highlight your skills and experience, show your portfolio, and set your ideal pay rate.
  • Talent comes to you icon
    Work the way you want
    Apply for jobs, create easy-to-by projects, or access exclusive opportunities that come to you.
  • Payment simplified icon
    Get paid securely
    From contract to payment, we help you work safely and get paid securely.
Want to get started? Create a profile

About Upwork

  • Rating is 4.9 out of 5.
    4.9/5
    (Average rating of clients by professionals)
  • G2 2021
    #1 freelance platform
  • 49,000+
    Signed contract every week
  • $2.3B
    Freelancers earned on Upwork in 2020

Find the best freelance jobs

Growing your career is as easy as creating a free profile and finding work like this that fits your skills.

Trusted by

  • Microsoft Logo
  • Airbnb Logo
  • Bissell Logo
  • GoDaddy Logo