Skip to main content

Complete our 3 minute feedback survey to help us improve the framework.

Machine learning engineer

Find out what a machine learning engineer in government does and the skills you need to do the role at each level.

Last updated 30 November 2024 — See all updates

What a machine learning engineer does

A machine learning engineer develops, assures and maintains machine learning models so they can be used in products and services.

In this role, you will:

  • be responsible for the software development and technical infrastructure needed to design, train, deploy and scale machine learning models
  • provide and maintain effective, secure and sustainable machine learning models for use in products and services
  • support all stages of the machine learning life cycle
  • help product teams evaluate and choose appropriate machine learning solutions

Machine learning engineer role levels

There are 2 machine learning engineer role levels, from senior machine learning engineer to lead machine learning engineer.

The typical responsibilities and skills for each role level are described in the sections below. You can use this to identify the skills you need to progress in your career, or simply to learn more about each role in the Government Digital and Data profession.

1. Senior machine learning engineer

A senior machine learning engineer develops machine learning models so they can be used in products and services.

At this role level, you will:

  • decide what model is most suitable for use in products and services
  • customise, optimise, re-train and maintain existing models
  • deploy models into production, testing and assuring them to ensure they meet performance requirements
  • work with others to integrate models with existing systems
  • check that models used in live products and services stay safe, secure and continue to work effectively

Skill Description

Applied maths, statistics and scientific practices

Level: practitioner

Practitioner is the third of 4 ascending skill levels

You can:

  • apply designated quantitative techniques such as time series analysis, optimisation and simulation to create and embed appropriate models for analysis and prediction
  • provide guidance on matching data sources with relevant applied mathematics and statistical techniques to meet analysis goals
  • apply appropriate statistical techniques to available data to discover new relations and offer insight into research problems, helping to improve organisational processes and support decision making
  • access and use the statistical tools available within the organisation

Communicating between the technical and non-technical

Level: practitioner

Practitioner is the third of 4 ascending skill levels

You can:

  • listen to the needs of technical and business stakeholders, and interpret them
  • effectively manage stakeholder expectations
  • manage active and reactive communication
  • support or host difficult discussions within the team or with diverse senior stakeholders

Data science innovation

Level: practitioner

Practitioner is the third of 4 ascending skill levels

You can:

  • demonstrate practical knowledge of data science tools and techniques
  • develop data science solutions that maximise insight
  • identify opportunities for how data science can improve data practices

Ethics and privacy (data science)

Level: practitioner

Practitioner is the third of 4 ascending skill levels

You can:

  • show an understanding of how ethical issues fit into a wider context and can work with relevant stakeholders
  • stay up to date with developments in data ethics standards and legislation frameworks, using these to improve processes in your work area
  • identify and respond to ethical concerns in your area of responsibility

Programming and build (software engineering)

Level: practitioner

Practitioner is the third of 4 ascending skill levels

You can:

  • collaborate with others when necessary to review specifications
  • use the agreed specifications to design, code, test and document programs or scripts of medium-to-high complexity, using the right standards and tools

Systems integration

Level: practitioner

Practitioner is the third of 4 ascending skill levels

You can:

  • define the integration build
  • co-ordinate build activities across systems
  • understand how to undertake and support integration testing activities

2. Lead machine learning engineer

A lead machine learning engineer leads the technical development and deployment of machine learning models.

At this role level, you will:

  • ead the most complex technical work needed to develop models for use in products and services
  • co-ordinate moving a model from the research and development stage to production
  • define ways of working across the machine learning life cycle
  • identify training needs for machine learning engineers and related roles
  • help your team work with other teams and disciplines
  • assure the effectiveness of machine learning models in use across the organisation
  • define and communicate software standards and guidelines related to ethics, risk and security
Skill Description

Applied maths, statistics and scientific practices

Level: practitioner

Practitioner is the third of 4 ascending skill levels

You can:

  • apply designated quantitative techniques such as time series analysis, optimisation and simulation to create and embed appropriate models for analysis and prediction
  • provide guidance on matching data sources with relevant applied mathematics and statistical techniques to meet analysis goals
  • apply appropriate statistical techniques to available data to discover new relations and offer insight into research problems, helping to improve organisational processes and support decision making
  • access and use the statistical tools available within the organisation

Communicating between the technical and non-technical

Level: expert

Expert is the fourth of 4 ascending skill levels

You can:

  • mediate between people and mend relationships, communicating with stakeholders at all levels
  • manage stakeholder expectations and moderate discussions about high risk and complexity, even within constrained timescales
  • speak on behalf of and represent the community to large audiences inside and outside of government

Data science innovation

Level: expert

Expert is the fourth of 4 ascending skill levels

You can:

  • be a leader in the data science space
  • demonstrate in-depth knowledge of data science tools and techniques, which you can use to solve problems creatively and to create opportunities for your team
  • act as a coach, inspiring curiosity and creativity in others
  • demonstrate in-depth knowledge of your chosen profession and keep up to date with changes in the industry
  • challenge the status quo and always look for ways to improve data science

Ethics and privacy (data science)

Level: practitioner

Practitioner is the third of 4 ascending skill levels

You can:

  • show an understanding of how ethical issues fit into a wider context and can work with relevant stakeholders
  • stay up to date with developments in data ethics standards and legislation frameworks, using these to improve processes in your work area
  • identify and respond to ethical concerns in your area of responsibility

Programming and build (software engineering)

Level: expert

Expert is the fourth of 4 ascending skill levels

You can:

  • advise on the right way to apply standards and methods to ensure compliance
  • maintain technical responsibility for all the stages and iterations of a software development project
  • provide technical advice to stakeholders and set the team-based standards for programming tools and techniques

Systems integration

Level: expert

Expert is the fourth of 4 ascending skill levels

You can:

  • establish standards and procedures across a service product life cycle, including the development product life cycle, and can ensure that practitioners adhere to these
  • manage resources to ensure that the systems integration function works effectively

Role Shared skills
Data scientist

Applied maths, statistics and scientific practices

Data science innovation

Ethics and privacy (data science)

Development operations (DevOps) engineer

Programming and build (software engineering)

Systems integration

Software developer

Programming and build (software engineering)

Systems integration

Analytics engineer

Communicating between the technical and non-technical

Data analyst

Communicating between the technical and non-technical

Updates

Published 30 August 2024

Last updated 30 November 2024

30 November 2024

  • The skill level descriptions for 'systems integration' have been updated to improve clarity, better meet definitions for skill levels and to ensure consistency across the framework. This allows the skill to be shared with roles previously using the skill 'systems integration (network architect)'.

30 August 2024

  • First published.