Join Halfspace’s AI & Advanced Analytics Team – Staff or Lead AI Safety Engineer
Accenture Nordics
At Halfspace, part of Faculty - we believe that the smallest edge can make the biggest difference. We began in sports, predicting outcomes in live football matches, and today we bring that same precision and ambition into the world of business. We're here to unlock value through artificial intelligence and help organisations make better, more informed decisions, faster.
With more than one hundred AI and data projects delivered, our work has transformed industries and earned international recognition. But awards and numbers tell only part of the story. What truly defines us is our people. Our team comes from leading institutions such as KU, DTU, Stanford, Oxford, Harvard, Caltech, Yale, Columbia, and many more, blending expertise in physics, engineering, mathematics, statistics, operations research, computer science, chemistry, economics, biology, design and strategy. Together, we share curiosity, boldness, and a pioneering spirit.
Halfspace is a part of Faculty, which forms the global spearhead for deep applied AI capabilities in Accenture. Faculty, founded in the UK and now expanding globally, is a leading European player in AI Safety, with a long history of working with Frontier AI Labs to evaluate frontier models.
What You'll Be Doing
You will be a key part of the team building Halfspace’s technical AI Safety capabilities. You will help define best practices for building safe and secure agentic AI systems, across rigorous evaluation, threat and risk modelling, control design, and robust system architectures. You will work closely with Halfspace leadership and commercial leads to shape our safety offering, and with senior client executives to scope engagements. You will have a lead role in Safety-focused client engagements, ensuring quality and rigour in our solution designs or assessments we do, while mentoring junior colleagues to deliver. You will work on both Downstream Safety, ensuring systems built around foundation models are safe, and Upstream Safety in cases where we evaluate models that clients train or fine-tune themselves.
We expect you to join at either Staff or Lead level, depending on your experience and qualifications.
You’ll work alongside excellent Halfspace engineers, many of whom are already delivering exciting Safety work, as well as engineers and Safety experts from Faculty and Accenture.
Who We're Looking For
We’re looking for engineers who have a proven track record of leading AI Safety work or adjacent fields, including e.g. evaluation design of agentic systems, AI red teaming, threat and risk modelling, control design and implementation. You understand the governance and regulatory requirements around this, but your core focus is to design and build systems to be safe and secure, and validate that they are. You may also have consulting experience and are comfortable with showing outcomes quickly in situations where you must drive that ambiguous requirements are translated into concrete, deliverable, and valuable technical solutions.
If you have experience in AI Safety but do not yet qualify for a Staff or Lead position, please apply to our AI Engineering position and flag your safety interest.
What You’ll Bring
These reflect the work you’ll be doing. Don’t hesitate to apply if you only meet some of them.