Job Description

Position Overview

This is a individual-contributor AI Engineer role for an experienced Machine Learning Engineer / Applied AI Engineer who has designed, deployed, and operated production machine learning systems in real-world, high-availability environments.

The role focuses on real-time inference, low-latency machine learning, production AI reliability, and end-to-end model ownership, where performance, stability, scalability, and controlled model evolution are more important than academic experimentation.

You will own the architecture, deployment, optimisation, monitoring, and lifecycle management of real-time AI models that operate during live voice and conversational workflows, influencing production decision-support systems.

This role is ideal for candidates searching for:

AI Engineer, Principal AI Engineer, Machine Learning Engineer, Applied ML Engineer, NLP Engineer, Real-Time ML Engineer, Inference Engineer, Production AI Engineer, MLOps-aware AI Engineer.

How We Work - Engineering Culture & Leadership

Culture is demonstrated through production ownership, operational excellence, and technical accountability, especially when systems are live.

All engineers operate

Above the Line

, meaning:

Accountability

- Own problems end-to-end, from detection to resolution

Respect

- Communicate clearly, professionally, and constructively

Action

- Drive progress independently in fast-moving environments

Feedback

- Seek continuous improvement through challenge and review

Ownership

- Take responsibility for system behaviour, not just code

Empathy

- Build AI systems that behave predictably for real users
At principal level, this means autonomous execution, sound technical judgement, and decision-making under real-world constraints.

What You Will Own

Applied Machine Learning & Production AI Systems

You will:

Architect, build, and operate production-grade machine learning systems Design applied ML models for continuous real-time inference Work with speech, audio, NLP, text, intent detection, sentiment analysis, and behavioural signals Optimise models for latency, throughput, cost efficiency, interpretability, and reliability Set technical direction for applied AI and ML approaches used in live systems
This role requires designing for edge cases, failure modes, degradation behaviour, and recovery, not just ideal conditions.

Real-Time Inference & Low-Latency Systems

You will:

Define and implement real-time inference strategies (event-driven inference, partial inputs, rolling context, sliding windows) Optimise ML models for low-latency, high-concurrency, predictable performance Collaborate closely with platform, infrastructure, and MLOps engineers Own decisions around model serving, batching, GPU utilisation, concurrency limits, and resource allocation Understand performance trade-offs across CPU vs GPU inference, throughput, memory, and degradation under load
Experience with real-time systems, streaming data, or live inference pipelines is critical.

Training, Evaluation & Continuous Model Improvement

You will:

Design and maintain training and retraining pipelines for structured and unstructured data Define model acceptance criteria, validation checks, and regression thresholds Establish meaningful offline and online model performance metrics Monitor and mitigate data drift, model drift, bias, and unintended behaviour Make informed decisions about when models should not be promoted or deployed
Ownership includes knowing when not to ship.

Model Lifecycle, MLOps & Production Governance

You will:

Own model versioning, metadata, and artifact management Define safe deployment strategies (canary releases, champion/challenger, staged rollouts) Monitor live model behaviour and intervene when performance or behaviour deviates Document assumptions, limitations, risks, and operational constraints Ensure models remain explainable, auditable, and production-safe
This role carries responsibility for production model behaviour, not just development output.

What This Role Is Not

This role is

not

:

A notebook-only experimentation role A junior, guided, or task-only position
This is a principal applied AI engineering role with real accountability for live systems.

Experience Required ( Principal Level)

You should bring:

Extensive experience shipping, deploying, and operating ML models in production Strong applied machine learning fundamentals and engineering judgement Expert Python skills and deep experience with modern ML frameworks (PyTorch, TensorFlow, scikit-learn, Hugging Face) Proven experience with real-time, streaming, or low-latency systems Confidence owning decisions that affect live system performance and behaviour
Highly desirable experience:

NLP, speech processing, conversational AI Inference optimisation, GPU-based serving, or high-throughput ML systems Applied experience at scale, not just prototypes
Job Types: Part-time, Temp to perm, Internship
Contract length: 12 months

Pay: 19,349.20-21,067.20 per year

Ability to commute/relocate:

Manchester M35 9BD: reliably commute or plan to relocate before starting work (required)
Experience:

AI: 3 years (required)
Work authorisation:

United Kingdom (required)
Work Location: In person