AI Platform Engineer - Insurance

Source Technology Limited; Greater London; On-site; GBP 90,000 - 140,000; Full time

Job summary

Source Technology Limited in London is seeking an AI Platform Engineer to design, build and operate a secure, scalable AI/ML platform for insurance use cases including pricing, claims automation, fraud detection and customer analytics.

You will partner with engineering, data, security and risk teams to enable reliable model development, deployment and monitoring in a regulated environment. Strong software and MLOps skills are essential.

Qualifications

• Strong software engineering in Python and/or Java/Scala.
• Hands-on ML platform experience with MLOps tools.
• Experience with Kubernetes, Docker, and IaC.
• Cloud experience (AWS/Azure/GCP) with secure architectures.
• Familiarity with data pipelines (Spark, Airflow, Kafka, dbt).

Responsibilities

• Design and maintain platform capabilities for model training and deployment.
• Build ML pipelines and CI/CD workflows for MLOps.
• Implement containerised deployments and orchestration for inference.
• Establish data and model governance and security controls.
• Provide developer tooling and documentation for data scientists.
• Monitor model performance and reliability with SLAs.

Skills

Python

Java/Scala

MLOps

Kubernetes

Docker

Terraform/CloudFormation

Cloud platforms

Data pipelines

REST/gRPC

PII security

RBAC

Tools

MLflow

Kubeflow

SageMaker

Vertex AI

Databricks

Terraform/CloudFormation

Job description

Location: London, United Kingdom

Industry: Insurance

Overview We are seeking an AI Platform Engineer to design, build and operate a secure, scalable AI/ML platform supporting insurance use cases including pricing, claims automation, fraud detection, customer analytics and document intelligence. You will partner with engineering, data, security and risk stakeholders to enable reliable model development, deployment and monitoring in a regulated environment.

Key Responsibilities

• Design and maintain platform capabilities for model training, feature engineering, deployment and observability across cloud and on-prem environments.
• Build CI/CD pipelines for ML (MLOps) including automated testing, packaging, model registry workflows and promotion across environments.
• Implement containerised deployments (Docker/Kubernetes) and orchestration for batch and real-time inference workloads.
• Establish data and model governance: lineage, reproducibility, access controls, retention, and auditability aligned with insurance compliance needs.
• Enable safe experimentation and cost control through resource management, autoscaling and quota policies.
• Implement monitoring for model performance, drift, bias and operational SLAs, with alerting and incident response practices.
• Harden platform security: secrets management, network policies, vulnerability management and secure-by-design patterns.
• Provide developer experience tooling, documentation and support for data scientists and software engineers.

Required Skills & Experience

• Strong software engineering skills in Python and/or Java/Scala, with clean coding, testing and code review practices.
• Hands-on experience with MLOps tools such as MLflow, Kubeflow, SageMaker, Vertex AI, Databricks, or equivalent.
• Proven experience with Kubernetes, Docker, Helm and infrastructure as code (Terraform/CloudFormation).
• Cloud experience with AWS, Azure or GCP (networking, IAM, storage, compute), including secure architecture principles.
• Experience with data platforms and pipelines (Spark, Airflow, Kafka, dbt or similar) and working with large-scale datasets.
• Knowledge of model serving patterns (REST/gRPC, batch scoring, streaming inference) and API gateway integration.
• Understanding of regulated data handling (PII), encryption in transit/at rest, and role-based access control.
• Ability to collaborate with risk, legal and compliance teams; familiarity with model risk management is advantageous.
• Excellent communication skills and ability to translate requirements into robust platform solutions.

Desirable

• Experience in insurance domain systems and data (policies, claims, underwriting, bordereaux).
• Knowledge of responsible AI practices, explainability tooling, and bias assessment.
• Experience with feature stores and governance tooling.

What You’ll Deliver

• A production-grade AI platform enabling faster, safer model delivery with measurable reliability and compliance.
• Standardised templates and pipelines reducing time-to-deploy and improving reproducibility.
• Operational metrics and monitoring that increase trust in models and platform performance.

We believe in equal opportunity for all and actively encourage applications from diverse backgrounds, experiences, and perspectives.