About the Role

Initial 12 month FTC with a view to going permanent.

About the Role

This is a hands-on AI engineering role focused primarily on building production-grade generative AI capabilities. We are looking for an experienced engineer with roughly 5-10 years of experience who has built real AI systems and wants to apply that foundation to modern GenAI use cases, including LLM access patterns, retrieval and RAG, tool integration, agent-style workflows, evaluation, observability, and guardrails. Experience with traditional ML or MLOps is helpful, but the primary emphasis is GenAI engineering and platform enablement.

You will help design, implement, and evolve GenAI platform capabilities that support practical enterprise use cases. This may include model access and serving, LLM gateways, retrieval and RAG, MCP and tool integration, agent patterns, evaluation, observability, monitoring, and AI guardrails. Where relevant, you may also help connect these capabilities with existing ML platforms or MLOps practices. You will work closely with colleagues across engineering, architecture, data, product, security, and business teams to turn real use cases into reliable platform capabilities.

This is not a pure architecture or strategy role. The role requires strong engineering judgment and growing architectural ownership, but the emphasis is on practical implementation: building components, validating trade-offs, integrating tools, and helping teams use the platform safely and effectively.

Responsibilities

• Build GenAI platform components: Develop production-grade services and integrations across LLM gateways, model access patterns, retrieval and RAG, MCP and tool integration, agent-style workflows, evaluation, observability, and guardrails.

• Translate use cases into architecture: Work with engineering, product, data, security, and business stakeholders to turn practical needs into clear technical designs and implementation plans.

• Evaluate technical trade-offs: Compare approaches across GenAI architectures, including hosted versus self-hosted models, gateway technologies, RAG patterns, agent and tool integration, evaluation strategies, and framework versus custom implementation choices.

• Integrate internal systems, data, and tools: Help connect approved data sources, platforms, APIs, models, and enterprise tools into AI-enabled workflows in a secure and maintainable way.

• Operationalize what you build: Design for authentication, authorization, rate limiting, cost tracking, monitoring, fallback, reliability, quality evaluation, and supportability from the start.

• Contribute to team direction: Bring technical depth, challenge assumptions, document decisions, and help shape the platform roadmap through hands-on learning.

Skills & Experience

We do not expect one person to be an expert across the entire AI ecosystem, but we are looking for strong hands-on depth in GenAI engineering, supported by solid platform engineering fundamentals:

• Typically 5-10 years of experience in software engineering, ML engineering, platform engineering, or a similar technical role.

• Hands-on experience building, deploying, or operating production GenAI or AI-enabled systems beyond prototypes and notebooks.

• Strong hands-on experience building with LLMs or GenAI systems, including model access patterns, prompt workflows, application integration, and production operational considerations.

• Experience with LLM gateways, API and service layers, or model routing patterns.

• Practical knowledge of RAG, embeddings, chunking, retrieval quality, grounding, and evaluation.

• Experience integrating AI capabilities with tools, APIs, enterprise systems, or agent-style workflows; MCP experience is a strong plus.

• Familiarity with hosted model APIs, self-hosted models, open-weight models, or traditional ML model deployment patterns.

• Strong Python, API design, CI/CD, cloud infrastructure, and production engineering fundamentals.

• Ability to explain technical trade-offs clearly to both engineers and non-technical stakeholders.

• Awareness of security, governance, monitoring, cost control, and quality guardrails needed for enterprise AI systems.

Nice to have

• Experience in regulated or highly governed environments.

• Knowledge of security, risk, and data governance practices.

• Experience with MLOps platforms, feature stores, model registries, or experiment tracking tools.

• Experience operating open-weight models such as vLLM, TGI, or Ollama.

• Experience with LLM observability platforms, ML monitoring, and evaluation pipelines.

• Analytics, data engineering, or applied data science experience.

• Multi-agent or advanced orchestration experience.

What we offer

We give you the freedom to be yourself. We are creating an environment of unique individuals – like you – with different perspectives on the financial industry and the world. A diverse and inclusive environment in which everyone’s voice counts and where you can reach your full potential.

We are committed to an inclusive and accessible recruitment process. If you require a reasonable accommodation related to accessibility during your application or interview, please contact [email protected] or indicate this in your application.

Please note that this mailbox is not monitored for general recruitment enquiries and should only be used for accessibility or accommodation-related requests (for example related to vision, hearing or neurodiversity).

All requests are confidential and will not affect your candidacy.

Don’t meet every single requirement? At Swift, we are dedicated to building a workplace where people can bring their full selves and ideas to the team, so if you are excited about this role, we encourage you to apply even if you do not meet every single qualification.