AI Engineer
Confirmed on the employer's own hiring board on Nov 2, 2026. First seen by Alion on May 21, 2026.
Job Title
London (On-site, London Farringdon)
Full-time, salary: £60,000-80,000
Job Summary
Travtus is looking for anAI Engineer to help design, build, and optimise the core AI platform capabilities that power agent-based products and workflows.
This role is focused on platform development. It is responsible for shaping how LLM-powered agents are designed, orchestrated, tested, monitored, and improved in production. The role sits at the intersection of prompt engineering, agent architecture, solution design, and platform capability development.
You will define how agents interact with tools, APIs, and downstream services, and help specify the requirements and contracts for the services they depend on. You will work closely with platform engineers, who build the APIs and services consumed by agents, and with product engineers, helping ensure agent capabilities support strong user experience and front-end design.
This role also includes working on patterns where prompt outputs are used as code to support interface generation, front-end behaviour, and rapid product iteration. You will also own the LLM and prompt tooling lifecycle, from prompt management through evaluation, testing, and continuous improvement.
This is a hands-on senior role for someone who combines deep technical skill with strong systems thinking and a practical approach to building scalable, usable, and production-ready AI capabilities.
About Travtus
Travtus is transforming the Housing Industry with our Everyday (AI)™ platform. We bring intelligence and automation together in one place so teams can move faster and work smarter. By delivering everyday AI decisions and workflows across the systems operators already use daily, our platform improves efficiency while enhancing the resident experience.
About the Role
Key Responsibilities
Design and develop multi-agent prompts and orchestration patterns
Define how agents, tools, APIs, services, and platform components interact, including requirements, contracts, and integration patterns
Work alongside platform engineers to define the APIs and services that agents consume and use
Work closely with product engineers to support user experience and front-end design, including the use of prompt outputs as code
Own LLM and prompt tooling across the full lifecycle, from prompt management and versioning through evaluation, testing, and optimisation
Design requirements for downstream and supporting services that enable agent execution, workflow control, observability, and integration
Lead the research, evaluation, and selection of frameworks for agent orchestration, tool use, evaluation, observability, and deployment
Perform model selection based on use case fit, quality, latency, scalability, and cost
Design and implement tools and function-calling patterns for agent-based systems
Carry out prompt tuning and iterative optimisation of agent behaviour and output quality
Monitor, debug, and continuously tune agents in development and production
Design for performance, scalability, reliability, usability, maintainability, and cost efficiency
Define and execute testing strategies for LLM agents and platform-level AI workflows
Integration test agents end to end with other parts of the platform to ensure reliability across workflows, services, and user-facing experiences
Build proofs of concept and prototypes to validate architectures, frameworks, and new platform capabilities
Produce end-to-end solution designs for AI platform components and services
Use AI to improve software development, including prototyping, implementation, testing, debugging, and engineering productivity
Collaborate across engineering and product teams to translate business and technical goals into robust platform capabilities
Requirements
The successful applicant will be a natural problem-solver with an affinity for automation, efficiency, and AI-driven solutions with excellent communication and collaboration skills across both technical and non-technical teams.
Must have requirements:
Strong experience designing, building, and deploying LLM-based applications or agentic systems
Demonstrated expertise in prompt engineering, prompt tuning, and structured output design
Experience designing multi-agent workflows and orchestration logic
Strong understanding of API design, service integration, and interface contracts
Experience owning or working with prompt management, evaluation, and testing tooling
Experience working with or defining requirements for platform services that support AI applications
Experience evaluating and selecting models, frameworks, and AI tooling
Practical experience with tool use, function calling, workflow control, and system integration
Experience defining and implementing testing and evaluation strategies for LLM systems
Experience with end-to-end and integration testing across complex platform workflows
Strong understanding of performance optimisation, including latency, throughput, scalability, and cost trade-offs
Experience monitoring and improving AI systems in production
Strong software engineering skills and experience building production-grade systems
Ability to work effectively with both product engineers and platform engineers
Ability to produce clear solution designs and communicate technical decisions effectively
Comfortable working across experimentation, architecture, and delivery
Nice to have requirements:
Experience with agent orchestration frameworks and evaluation platforms
Familiarity with observability, tracing, and monitoring for AI systems
Experience designing guardrails, controls, and failure handling for AI workflows
Experience supporting front-end or UX-oriented engineering through AI-assisted design or generation patterns
Experience with code generation or structured prompt outputs used in software delivery workflows
Experience with cloud-native or distributed production environments
Background in solution architecture, or applied AI systems
Experience using AI to improve internal engineering productivity and delivery
What Success Looks Like
The AI platform provides a strong, scalable foundation for agent-based products and workflows
Agents, tools, APIs, and service integrations are reliable, testable, and production-ready
LLM and prompt tooling supports disciplined prompt management, evaluation, testing, and continuous improvement
Platform capabilities are designed with strong usability, maintainability, and operational robustness
Agent outputs support high-quality product experiences, including front-end and UX use cases where appropriate
Model, prompt, and framework choices are evidence-based and continuously optimised
Agents are integration tested end to end with the wider platform and perform reliably across real workflows
New ideas can be rapidly validated through POCs and evolved into durable platform capabilities
Product engineers and platform engineers can work effectively around clear contracts, services, and orchestration patterns
AI is used effectively both within the product and to improve software development practices
About the Team
We are a dedicated team, working in a truly collaborative style, where everyone is heard and brings something valuable to the conversation. Constantly pushing boundaries to create groundbreaking solutions for our customers. We are fundamentally challenging the way one of the largest industries in the world operates, and our commercial success pays testament to the skill, commitment and passion that our team displays every day.
Working together in person fuels our best ideas and innovations, which is why all team members spend at least three days a week in our London office.
Benefits
Be part of a growing, mission-driven startup with global ambitions
Exposure to cutting-edge AI technology in a practical, high-demand sector
Collaborative, fast-paced team culture with space to innovate and grow
Full-time, salary: £60,000-80,000
On-site working model with a modern London office (Farringdon)
Private Healthcare
Pension
Deliveroo Allowance
Help us shape the future of technology. Apply Now!