Principal AI Platform Engineer (Python)

Our client is a leading global investment management company headquartered in London. It manages over $228 billion in assets and serves institutional investors, pension funds, wealth managers, and other sophisticated clients worldwide. The firm specializes in quantitative investing, alternative investments, systematic trading strategies, and technology-driven asset management. Data science, machine learning, and AI are core components of its investment and research processes.

As part of our collaboration we will focus on two foundational capabilities required to enable safe and scalable AI adoption across the enterprise: Agentic Security and AI-Ready Data Foundations.

We build the data foundations and evaluation frameworks that make AI useful, reliable and safe inside regulated financial firms. The value of an AI agent depends not only on the models behind it, but also on the quality of the structured and unstructured data it consumes and the accuracy, relevance and traceability of the outputs it produces. Your job is to measure that quality, identify where it breaks down and turn the findings into practical improvements.

This is an engineering role, not an analytical one. You will build the agentic workflows that reason over the firm's research content, and the ingestion, evaluation and guardrail tooling that makes their output trustworthy enough for investment professionals to act on. None of this tooling exists today — you would be building it from scratch.

Requirements:

• 7+ years of software engineering experience, a substantial part of it spent building infrastructure or platform tooling rather than operating it
• Python (required): production-level Python, including packaging, typing, testing, and dependency management. The platform is written in Python and you will be extending it, not scripting around it. Bash for glue work.
• On-premises infrastructure: solid Linux fundamentals, and working knowledge of the networking, storage, and virtualization layers the platform sits on. Platform code talks to this hardware directly, so you need to understand what is underneath the abstraction.
• Infrastructure as code: hands-on experience with Terraform and Ansible, which are the tools in use here. Infrastructure here lives in version control and goes through review like application code.
• Containers and orchestration: Docker and Kubernetes, including how workloads are scheduled, configured, and given access to resources at runtime.
• CI/CD and DevOps practice: experience building pipelines (GitLab CI, GitHub Actions, Jenkins) that test, package, and release software, and the practices around them, such as automated testing and staged rollouts.
• Observability: experience instrumenting systems with Prometheus, Grafana, OpenTelemetry, or an equivalent stack, and using that data to diagnose failures in distributed systems.
• Security (critical): a working grasp of secrets management, identity and access control, network isolation, supply-chain and dependency risk, and how to build these into a platform so that consuming teams inherit them by default.
• Collaboration and communication: you can gather requirements from engineering teams and write documentation others can work from. You will also need to explain design decisions to people outside engineering.
• Problem-solving: the judgment to work through system failures that have no obvious cause, and the patience to find out why rather than restart the service.

Personal Skills

• Adaptability: comfortable when priorities, tools, or requirements change mid-project.
• Curiosity: you follow how other teams solve infrastructure problems and bring back what works.
• Communication: you write and speak clearly enough that other teams can act on it without a follow-up meeting.
• Teamwork: you work well with people outside your own team and share what you know.
• Ownership: you follow a problem to its cause rather than handing it on when the symptom clears.
• Attention to detail: you notice the small things, like an unpinned dependency or a permission wider than it needs to be.
• Prioritization: you can judge what to do first when several teams need something at once.
• User focus: you treat the engineers who use the platform as your users, and judge your work by whether theirs gets easier.

Responsibilities:

• Designing, building, and maintaining the Python services, libraries, and command-line tooling that make up the internal infrastructure platform.
• Working with the engineering teams who consume the platform to understand their workflows and turn recurring pain points into platform features.
• Replacing manual infrastructure operations with codified, self-service workflows, so that environments are reproducible, reviewable, and safe to change.
• Building observability into the platform through metrics, structured logging, and alerting, so failures surface before consuming teams report them.
• Owning production issues in the platform end to end, from triage and root cause analysis through the code or configuration change that prevents a recurrence.
• Managing and scaling the compute and storage the platform provisions across on-premises hardware, balancing performance against cost as usage grows.
• Building security into the platform by default: secrets management, least-privilege access, dependency hygiene, and an auditable history of every infrastructure change.
• Evaluating new tools, libraries, and patterns, and folding the ones that earn their place into the platform roadmap.

Why this position:

This role sits at the intersection of data engineering, AI, and financial services, solving one of the most important challenges in enterprise AI: enabling agents to securely access and reason over trusted data. You'll have the opportunity to design and build foundational platforms that combine large-scale data systems, governance, and AI technologies in highly regulated environments. It offers significant technical ownership, exposure to cutting-edge AI agent architectures, and the chance to shape how organisations safely unlock value from their data.

#J-18808-Ljbffr