← Live feed

Developer Wanted: Simple, Private Offline AI Image System

Budget: - HOURLY / PART_TIME ⭐ 0.00 (0) CAN

python, computer-vision

Preferred qualifications

  • Experience: Expert
I am looking for an experienced developer to build a reliable, easy-to-use AI image system that runs locally on my Windows computer. The process must be simple: upload images, enter a prompt, choose how many variations I want, and click Create. No manual masking, node editing or complicated controls. The system must do three things: Replace characters in an existing image: Upload a scene and character reference images, then replace the people in the scene while preserving its setting and composition. Create scenes using uploaded characters: Upload images of one or more characters, describe a scene in plain English, and generate images featuring those characters. Edit a character’s appearance: Upload a character image and change their hair, clothing, weight, height or body proportions while preserving their identity and any features I have not asked to change. Image quality and character accuracy Results must be highly photorealistic. Facial likeness and body proportions must match the uploaded references as closely as possible—ideally approaching 100%—except where I request changes. Consistent identity across different scenes is essential. Please demonstrate achievable accuracy using my reference images and clearly explain any limitations. Ease of use and privacy One simple interface and one launcher. No manual technical steps during normal use. Generate as many variations as I choose, queued in batches if necessary, without an arbitrary output cap. Visible progress, image previews, cancellation and straightforward saving. Fully offline operation after installation, with images and prompts remaining on my computer. Clear error messages if something fails. My computer Windows PC AMD Ryzen 7 9700F, 8-core processor 32 GB RAM NVIDIA GeForce RTX 5070 with 12 GB dedicated VRAM Approximately 2 TB storage, with 1.49 TB currently free ComfyUI, LM Studio and a custom interface called Verbosity are already installed. You may reuse suitable components or propose a replacement. Project delivery Previous attempts have consumed months without delivering the required results. I need someone who has already built comparable systems and can deliver a finished, usable product. I prefer a fixed-price project with clear milestones. The first milestone must demonstrate all three functions on my computer using images and prompts I provide, before I commit to the full build. Final delivery must include the full source code for custom work, required workflows and model details, installation/recovery instructions, a short user guide and an agreed bug-fix support period. Please include in your proposal: Examples of similar systems you have personally delivered. Your proposed approach and its limitations. Realistic image quality, resolution and generation times on my hardware. Any required hardware upgrades and their estimated costs. Your fixed-price milestones, delivery timeline and communication availability.
Open job

AI proposal draft

Generate a short cover letter to copy into the offer. Says you are interested and ready to work.

Sign in to generate an AI proposal draft.

Log in