Senior AI Engineer (Forecasting & Anomaly Detection)

Strategic Blue builds products that help organisations understand, forecast and optimise their AWS spend. We're looking for a Senior AI Engineer to take technical ownership of the machine-learning capabilities behind our forecasting and anomaly-detection products.

This is a hands-on, high-autonomy role for someone who enjoys solving complex applied ML problems.

Initially, you'll focus on forecasting, anomaly detection and integrated user feedback - including how customers can correct or steer forecasts and how our models should learn from those interventions.

You'll work closely with our Product team, turning product ideas into robust technical solutions.

We'll give you significant autonomy over the approach: we want someone who can explore different algorithmic options, evaluate them rigorously, make sound engineering decisions and clearly explain the trade-offs behind them.

• Forecasting & anomaly detection - owning and improving our existing models and developing new algorithmic capabilities.
• Feedback-driven forecasting - developing ways for customers to adjust and influence forecasts, and for the system to intelligently learn from that feedback.
• Model evaluation - designing robust benchmarks, backtests and metrics, including segment-level analysis and safeguards against issues such as lookahead leakage.
• Algorithmic alerting - determining when alerts should fire, appropriate thresholds and how suppression logic should behave.
• The model-facing data layer - working with our internal analytics API and data pipelines across FastAPI, Athena and Postgres.
• Technical decision-making - evaluating different approaches, documenting your reasoning and communicating recommendations clearly to both technical and non-technical colleagues.

Our existing development team owns general backend operations and data refresh. You'll work closely with them to define what the models need and ensure your work integrates effectively with our existing systems.

Essential

•      Traditional ML expertise. Including statistical expertise to interpret and improve upon existing model performance, not just maintaining the Python system around them. Some LLM or agentic expertise desirable for planned features.

•      Strong Python, and the ability to work in a FastAPI / Athena / Postgres stack.

•      Fluent, critical use of AI coding agents. We expect you to work with AI tooling — but not to vibe-code. Pushing verbose AI-generated code and hoping is a dealbreaker.

•      Translation between technical and product. You can turn product intent into a specification, and explain a trade-off to someone who does not model for a living.

•      Appetite for AWS billing data. A critical part of the job is mastering the nuances of AWS billing data. CUR, Savings Plans, marketplace charges and account structures are uniquely intricate. If this sort of problem piques your curiosity, we’re keen to hear your approach.

Nice to have

•      FinOps or AWS billing experience. The intersection of ML and FinOps is small, so this is a bonus rather than a filter.

•      basic React – supporting the process from data science prototype to client interface.

This is a contract position for 3–5 days per week, with flexibility for remote and hybrid working from London.

The work has previously been delivered successfully across three days per week, although we're open to increasing this depending on experience, availability and delivery timelines.

We're looking for someone who can operate as a senior technical expert rather than simply execute a predefined backlog. You'll have the freedom to challenge an approach, propose something better and explain why.

Apply here and send us a short email or even better, a quick video introducing yourself. We'd particularly like to hear about a forecasting or anomaly-detection system you've personally owned. Tell us what you built, what you got wrong the first time, and what you learned or changed as a result.