We are building an AI-powered Meal Scan system that analyzes meal photos and estimates ingredients, portion sizes, food volume, calories, macronutrients, micronutrients, and personalized nutrition insights. We are looking for a senior AI Engineer who can take ownership of the entire AI stack, including research, prototyping, model selection, evaluation, deployment, and optimization. The ideal candidate has hands-on experience with:

• Computer Vision (food recognition, object detection, segmentation)
• Vision Language Models (Qwen, Qwen-VL, Llama, Gemma, etc.)
• vLLM deployment and optimization
• Multimodal AI pipelines
• Portion size and volume estimation from images
• Depth estimation and 3D reconstruction techniques
• PyTorch and Hugging Face ecosystem
• Fine-tuning and LoRA training
• MLOps and production AI systems
• FastAPI, Docker, Kubernetes
• AWS or GCP infrastructure

Responsibilities include:

• Improving food and ingredient recognition accuracy
• Developing volume and portion size estimation algorithms
• Evaluating and integrating state-of-the-art VLMs
• Building scalable inference pipelines using vLLM
• Designing evaluation frameworks and benchmarking systems
• Deploying and maintaining production AI services
• Collaborating with nutrition, product, and engineering teams

We are looking for someone who can work independently, propose technical solutions, make architecture decisions, and drive the AI roadmap with minimal supervision. Experience in food AI, nutrition applications, healthcare AI, or multimodal production systems is a strong plus.