Gen-AI Engineer (Image Generation)

About the company :

• At Alle, we are building an AI-powered personal fashion assistant.

• The assistant helps users discover, try-on & shop the best fashion trends.

• We are funded by top VCs and angels from India & US.

• You'll be part of the founding team and work closely with founders.

• This is a great opportunity if you're interested in building a global consumer AI product from India.

• Download our app: Android, iOS

• You can also check our careers page.

Founding team :

All three of us were part of Meesho's founding team where we spent 5+ years building the largest fashion e-commerce of India.

• Harshit Madan - 1st Product Leader at Meesho, ex-founder, IIT KGP

• Pavan Patil - 1st Backend Engineer at Meesho, ex-founder, IIT Roorkee

• Prateek Agarwal - Founding team at Meesho, ex-founder, IIT KGP

All of us were founders before joining Meesho and we love hiring people who have a founder mindset.

What will you be working on?

• Build end-to-end workflows for fashion image generation ensuring consistency in outfit styles, textures, and aesthetics.

• Research and implement techniques to improve the realism and detail of AI-generated fashion images, such as achieving perfect poses, garment textures, depth of field etc.

• Train and fine-tune diffusion (transformer) models and LoRAs for generating fashion images.

• Optimize models for efficiency, speed, and quality while balancing compute costs and visual output.

• Establish rigorous evaluation metrics to assess the realism, diversity, and coherence of AI-generated fashion images.

Criteria

• Strong knowledge of Flux/Stable Diffusion workflows, including: Text-to-image, Image-to-image, Inpainting, ControlNet, Face swap & other relevant techniques

• Understanding of key model components like text encoders, samplers, VAEs, etc.

• Hands-on experience with ComfyUI (mandatory) and other popular image generation tools

• Ability to fine-tune image generation models using techniques like LoRA. Having public projects on CivitAI, Hugging Face, or GitHub is a plus.

• Active in Stable Diffusion/Flux communities (Discord, Reddit, etc.).

• Portfolio or proof of work in AI-generated content.

How we hire?

• If you have 2-6 years of experience in computer vision or have build s with Gen-AI, please fill the form.

• We check every application in 72 hours and respond irrespective of our decision.

• If we like your work, we are quick to get on calls and roll-out an offer.

• Prior experience in startups and/or large-scale B2C companies is a big plus.