Lead MLOps Engineer, Quantexa

• Take pride in designing, productionising and deploying high quality, well-engineered, Machine Learning Training and Inference Pipelines and tools to improve the efficiency of Data Scientists
• Experience of Leading a team and mentoring junior engineers
• Guiding team members through technical issues and challenges
• Ensuring that engineers are following best practices and standards
• Leading SCRUM ceremonies and fostering a collaborative and fun working environment that values high performance and continuous improvement
• Great MLOps development experience:
• Write defensive, fault tolerant, efficient and well-engineered code following best software engineering practices
• Experience in code profiling and low-level optimization
• Good knowledge of a range of ML Ops tools such as MlFlow, Kubeflow, DVC or Weights and Biases
• Knowledge of the technologies used within the Quantexa platform and especially ML components within it:
• Container architectures, especially Kubernetes, Docker, and related technologies such as Terraform and Helm
• Dev and ML Ops tooling/automation written with Bash, Python, Jenkins and Groovy
• Python and the most established ML Libraries such as pytorch and sklearn
• Cloud platforms (GCP, Azure, AWS)
• Fascination with emerging technology and a passion for improving Machine Learning solutions

Desirable

• As part of our ongoing growth we’re standing up an ML Ops team to more efficiently build, maintain and deploy the increasing number of AI models we provide with our platform
• We’re recruiting the Team Lead to establish and run this new team
• You’ll be supporting our two Data Scientist teams who create the generic Machine Learning Models and Data Science based components for use in our solutions
• Our Analytical Innovation Team focuses on structured data, especially graph-based models and detection of risk
• Our NLP group teaches machines to understand natural language, building products to help extract meaning and insight from text, while also conducting exploratory research that we believe will drive improvements in our products and advance the state-of-the-art
• To make it faster, and more predictable for both Quantexa’s in-house Data science teams to build, deploy, monitor and maintain the various types of Machine Learning model we are creating
• Make deployment of models on site and in cloud simpler, faster and more consistent
• Where possible, and appropriate, the tools created should be extensible for use by clients for their custom ML models based on Quantexa Platform
• Establish the Quantexa ML Ops team and the team’s interfaces to other parts of Quantexa
• Advance our approaches to deployment of ML models within our platform
• Work with Architects and DS teams to select appropriate model architectures and patterns, including detailed dependencies. Optimise models to meet non-functional requirements. Document new approaches for reuse
• Building tools and processes to stand up required modelling environments on cloud, including environments with specialised capabilities such as GPUs
• Compose frameworks to support the Deployment, Testing, Monitoring and Governance of models
• Input into selection of 3rd party products and tools, as required, including any Vendor selection
• Provide support for teams using ML models. Examples include liaising with cloud and DS teams to resolve issues with live models, supporting client deployment of models in partnership with DS teams; Providing Model API support to Quantexa’s platform teams