Senior Machine Learning Engineer - Scan, Match and Catalog

About the role

We are seeking a Machine Learning Software Engineer to join Fetch's Scan, Match & Catalog team. This role sits at the intersection of applied machine learning, data engineering, and production systems, with a focus on improving receipt understanding, product matching, and catalog enrichment at scale.

Responsibilities

• Build and scale ML models across the scan, match, and catalog pipeline, supporting receipt understanding, product matching, and catalog enrichment.
• Implement and iterate on active learning strategies, including data sampling, error-driven retraining, and human-in-the-loop workflows.
• Leverage LLMs to reduce model training and annotation effort, including synthetic data generation, assisted labeling, weak supervision, and error analysis.
• Own ML experimentation, evaluation, and production inference for assigned SMaC components.
• Collaborate with product, data, and platform partners to translate quality gaps into ML improvements.
• Use AI tools to accelerate development and improve system design, including: Prototyping and validating ideas with LLM tools, leveraging AI for code iteration and experimentation, using AI assistants for architecture diagramming and design validation, exploring LLM-powered features where appropriate.

Requirements

• 4+ years experience in software engineering, with production-level coding experience.
• Strong proficiency in Python for ML development, with working knowledge of Go, and hands-on experience deploying models into production systems.
• Experience with AWS technologies and distributed systems.
• Practical experience applying LLMs to reduce training and annotation effort, including assisted labeling, synthetic data generation, weak supervision, or error analysis.
• Strong engineering mindset with the ability to deliver reliable, maintainable, and scalable systems.
• Experience with AI-assisted coding tools (e.g., GitHub Copilot, ChatGPT, or similar) to improve development efficiency and code quality.
• Able to critically evaluate AI-generated outputs, with strong debugging and problem-solving skills to validate correctness.