Company Description

DRESSX is a pioneer in AI-powered solutions designed for fashion ecommerce, specializing in transforming static product pages into interactive shopping experiences. By offering Virtual Try-On capabilities, AI-driven sizing recommendations, Mix & Match styling, and AI-generated content, DRESSX empowers fashion brands to enhance customer engagement and streamline ecommerce operations. With an industry-specific AI Suite, our technology integrates seamlessly into existing systems, enabling brands to generate photorealistic visuals, provide personalized shopping journeys, and drive measurable business outcomes. Trusted by leading fashion brands, DRESSX elevates customer shopping confidence while reducing returns and increasing conversion rates. Our solutions align with the evolving demands of today’s retail landscape, where personalized and interactive experiences are key to success.

Role Description

We are looking for a Computer Vision / Generative AI Engineer with strong hands-on experience in generative image models and production-grade computer vision pipelines.

The candidate should be able to work deeply with modern image generation models, understand how they work under the hood, adapt them to specific product needs, and turn research or experimental ideas into reliable working systems.

Key responsibilities

GenAI R&D: Adapt, fine-tune, and deploy FLUX and diffusion models.
CV Pipelines: Build controlled generation workflows combining GenAI with detection and segmentation.
Paper-to-Code: Implement cutting-edge research into production-ready features.
Inference Optimization: Speed up models and reduce GPU costs (quantization, compilation, caching).
Data Prep: Collect, filter, and annotate high-quality datasets for training.
API & Deployment: Containerize ML services (Docker) and build scalable APIs.

Key skills and experience

Strong understanding of modern transformer-based image generation architectures, especially FLUX-based models, including their architecture, training logic, inference process, and practical limitations.
Ability to build custom inference pipelines on top of base generative models and adapt them to specific product or business requirements.
Experience building end-to-end computer vision pipelines: reading research papers, documenting findings, collecting and preparing datasets, training models, validating results, and iterating on model quality.
Understanding of inference optimization concepts such as quantization, compilation, and caching.
Practical experience with Python, PyTorch, Diffusers, and OpenCV.
Experience packaging ML/CV solutions into Docker containers and exposing them through APIs for testing and production use.
Experience working with modern image generation APIs is a strong plus, including Nano Banana and GPT Image.
Solid general computer vision background and familiarity with common CV tasks such as detection, segmentation and classification.