Lead Machine Learning Engineer

🎯 What You’ll Be Doing

• Lead the end-to-end development and scaling of our core photo logging AI product, ensuring high availability, performance, and reliability for millions of users.
• Design, build, and refactor production-grade ML codebases, applying advanced software engineering principles (e.g., modular design, clean architecture, testability, dependency management) to create robust and maintainable systems.
• Deeply understand, implement, and improve LLM based AI products, including fine-tuning embedding models for semantic search, advanced prompt engineering techniques, and efficient context management strategies.
• Diagnose, debug, and resolve complex issues within ML pipelines, including performance bottlenecks, data quality problems, and model accuracy.
• Drive MLOps practices for model deployment, versioning, monitoring, and A/B testing, ensuring seamless integration and continuous improvement in production.
• Collaborate closely with data scientists to transition models into production, with platform engineers on infrastructure and scalability challenges, and with product/design teams to translate user needs into technical solutions.
• Stay at the forefront of ML research and technologies, actively evaluating and integrating cutting-edge advancements (especially in LLMs, multimodal AI) into our product and engineering workflows.
• Write clean, efficient, and well-documented code across our microservices and data pipelines, contributing to a culture of engineering excellence.
• Ship high-quality code to production frequently, ideally on a daily basis.

🧠 What You’ll Bring to the Table

• 7+ years of professional experience in backend software engineering, with at least 5 years explicitly focused on the full lifecycle of deployed Machine Learning or AI products.
• Proven expertise in designing, building, and extensively refactoring production-grade ML codebases, demonstrating strong command of software engineering best practices for complex, stateful systems.
• Deep practical expertise working with productionised LLM based products
• Experience with vector databases and similarity search.
• Hands-on experience with embedding models and their fine-tuning for semantic search.
• Prompt engineering.
• Demonstrated ability to diagnose and troubleshoot complex issues within ML pipelines, including performance, memory management, and data-related challenges.
• Strong command of Pythonor Java
• Extensive experience working on large-scale backend systems, dealing with scaling challenges, high availability, and supporting a high volume of users in an ML context.
• Hands-on experience with MLOps practices, including model deployment, monitoring (e.g., drift, quality), versioning, and A/B testing ML systems in production.
• Proficiency with cloud platforms such as Google Cloud (preferred, including Vertex AI) or AWS (including SageMaker).

🏆 Bonus Points - Not required, but WOW us!

• Experience with multimodal AI, particularly in areas related to image/vision and text fusion.
• Contributions to open-source ML projects, research papers, or conference presentations.
• Experience with ML frameworks like TensorFlow, PyTorch, or scikit-learn.
• Prior experience in data analysis or data science, providing a holistic understanding of the data lifecycle.

✨ Our Hiring Journey

• 👋 Talent Acquisition Screen – 40 mins A friendly chat with a member of our Talent team to understand your experience, career aspirations, and motivations for joining ZOE. This is also a great opportunity for you to ask any initial questions about the role or our culture.
• 🧠 Hiring Manager Screen – 30 mins A focused discussion with the hiring manager to delve deeper into your past leadership experiences, your approach to building and scaling ML products, and your initial thoughts on the challenges and opportunities at ZOE.
• 📊 Live ML Engineering Coding Interview – 60 mins This hands-on session will assess your practical ML engineering skills. You'll work with an existing Retrieval-Augmented Generation (RAG) codebase, focusing on analysis, debugging, and strategic refactoring for production. We encourage and expect the thoughtful use of AI assistants (like GitHub Copilot, ChatGPT, Gemini, etc.) during this interview, reflecting our day-to-day embrace of these tools. Be prepared to articulate your choices and the underlying ML and software engineering principles.
• 🛠️ System Design Interview – 60 mins In this session, you'll be given a high-level problem statement and asked to design a scalable, maintainable, and efficient Machine Learning system. We'll explore your architectural thinking, MLOps knowledge, and ability to justify design decisions with trade-offs.
• 💬 Final Round Interview (Behavioural & Leadership) – 45-60 mins This final conversation will focus on your leadership philosophy, how you navigate complex situations, foster collaboration, and embody ZOE's values. We'll explore your problem-solving approach in non-technical scenarios and your ability to drive team success.
• 🎉 Offer Stage We like to move fast, keep things clear, and make sure you feel supported every step of the way! If you're the right fit, we'll extend an offer and guide you through the next exciting steps.
At ZOE, we believe in a transparent and collaborative hiring process that helps us get to know you and allows you to learn about us.