Careers

AI Engineer

Experienced

Data

United Kingdom

An AI Engineer at Valcon combines software engineering with applied data science to build AI systems people can rely on.

This role focuses on hands-on development of applications with large language models at their core, supported by sound data analysis, experimentation and evaluation. You work directly with clients to identify useful opportunities and turn them into working solutions.

We bring a forward-deployed mindset – getting close to users, understanding their work and taking responsibility for useful outcomes. We use the title AI Engineer to reflect our technology-agnostic consulting remit. You may work with Databricks, Azure, AWS or another hyper-scaler on a variation of stacks, guided by the client’s needs and environment. You should be willing to learn new platforms and apply your engineering judgement across them.

This is a T-shaped role with depth in technical delivery and breadth across consulting, business processes and change. Some engagements focus on discovery and use-case assessment; others on prototyping and production delivery. You should enjoy contributing across that mix, recognising when greater impact depends on changes to workflows, data or operating models, and involving specialists where needed.

Experience

Strong engineering skills: writing maintainable code, structuring applications, integrating APIs, debugging failures and using source control and automated tests. You can take an experiment into a usable application.
Hands-on experience building applications with LLMs. You understand model APIs, structured outputs, context engineering, retrieval-augmented generation and tool calling, and can explain the trade-offs behind your design choices.
Applied data analysis using Python and SQL, including exploratory analysis, data preparation and quality checks. You can identify gaps or bias in the available data and judge whether it is suitable for the task.
• Experience applying statistical modelling or machine learning to a practical problem. You understand appropriate validation, training and test separation, data leakage, overfitting and the limitations of performance metrics.
• Experience designing experiments and evaluations: representative test data, useful baselines, error analysis and human review. You can explain how you decided whether a model or AI application was good enough for its intended use.
• Practical experience deploying or integrating software in a cloud environment. You understand security, data access, CI/CD and operational monitoring, and can adapt these practices to the client’s stack.

What you will do

Shadow users and map their activities, hand-offs and exceptions. Identify where AI can reduce repetitive work, improve decisions and give people more time for judgement and creativity.
Assess opportunities against value, feasibility, data availability and risk. Explain the case for investment, distinguish quick wins from larger changes and agree how success will be measured.
Explore and prepare data, investigate patterns and test hypotheses. Build the datasets and pipelines needed for experiments, working with data engineers and domain specialists where appropriate.
Build working prototypes quickly and test them with users. Develop applications (Python preferable) that connect LLMs to relevant context, tools and business systems, choosing agentic or conventional approaches according to the problem.
Apply statistical modelling or machine learning where it helps answer the question. Establish baselines, validate results and communicate what the evidence supports, including uncertainty and limitations.
Engineer reliable delivery through automated tests, AI evaluations, deployment pipelines and observability. Investigate failures across models, retrieval, tools and integrations, balancing quality with latency and cost.
Help clients put solutions into use with appropriate permissions, human review, failure recovery and ongoing ownership. Support changes to workflows, roles and controls, and check whether the expected benefits materialise.
Build trust through clear advice and working software. Contribute to proposals and demonstrations, share lessons and develop reusable engineering patterns with the wider team.

Consulting and ways of working

You listen carefully, lead useful discovery conversations and translate an ambiguous business problem into a practical experiment or delivery plan. You can explain technical choices clearly to users and senior stakeholders.
You understand that adoption depends on the surrounding process, skills, controls and ownership. You are willing to contribute to discovery, business cases and change as well as hands-on engineering.

We expect working capability across the core areas above, with deeper expertise in at least one. Your strongest background may be software engineering or applied data science. You should be able to explain your contribution to a solution, what failed and how you improved it, and know when to draw on other specialists.

Additional strengths

Deeper experience in an area such as forecasting, optimisation, computer vision, model adaptation or agent orchestration is welcome. Other useful strengths include MCP integrations, operating production AI systems and using AI coding tools with effective review and testing. These are examples of specialisms you can bring; expertise in every area is not expected.

The opportunity

You will help shape problems as well as build solutions, working with users whose needs become clearer through experimentation. You can influence technical choices, see how your work performs in practice and help develop Valcon’s reusable AI capabilities and engineering approach.