Join Multiverse as a Staff AI Engineer to secure AI systems and enhance workplace tools in a hybrid work environment.
Posted by employer 13 hours ago
First seen on Joblaze 27 minutes ago
Last verified on the company career page 27 minutes ago
Skills & Technologies
What you'll build
Must have
Nice to have
Practical constraints
AI in the day-to-day
Mae, our AI guide inside Mastry, is a real-time AI agent that works alongside the tools you already use.
Not disclosed in this posting: compensation, years of experience, visa sponsorship.
Benefits
Joblaze summary
In this role, the Staff AI Engineer focuses on integrating security measures into the production path and ensuring the safety of AI agents and workplace tools. Key skills include security engineering, production AI agent development, and a full-stack delivery approach, with an emphasis on creating user-friendly solutions. This position is ideal for someone with a strong background in security systems and AI, capable of influencing technical direction and collaborating across teams. Multiverse IO is scaling rapidly, aiming to enhance its AI capabilities while maintaining a secure environment.
Joblaze insights
Quick facts
From the original posting
Multiverse is the upskilling platform for AI and Technology adoption.
We have partnered with 1,500+ companies to deliver a new kind of learning that's transforming today's workforce.
Our upskilling apprenticeships are designed for people of any age and career stage to build critical AI, data, and tech skills. Our learners have driven $2bn+ ROI for their employers, using the skills they've learned to improve productivity and measurable performance.
In April 2026, we announced $70 million in strategic funding, led by Schroders Capital, with participation from StepStone Group, Lightspeed Venture Partners and General Catalyst. At an increased valuation of $2.1bn, the round makes us Europe's first EdTech double unicorn.
But we aren't stopping there. With a strong operational footprint and 800+ employees, we have ambitious plans to continue scaling. We're building a world where tech skills unlock people's potential and output.
Join Multiverse and power our mission to equip the workforce to win in the AI era.
The Agentic Work Environment (AWE) owns the environment everyone at Multiverse works in, wherever they are in the world. That covers the devices, workplace tools and accounts colleagues use every day, the platform our product is built and run on, and the identity and access that decide who, or which agent, can do what across all of it. Product teams own what they build; we own the ground it stands on, the path it takes to production, and the baseline standards it meets on the way. Our mission is to make the safe way the fast way for every colleague and the agents they put to work.
For a security-minded AI engineer, that remit runs in two directions. In the product platform, you build security into the path itself: review agents that check every change against our baseline standards, one identity for customers, staff and agents, and detection and response through our SIEM and SOAR, so nobody, person or agent, has to remember to be secure. In the workplace, you secure the devices, tools and accounts colleagues around the world rely on. As more of them put agents to work, whatever their job title, you make sure every agent has its own identity, a named owner, scoped access and evals, so autonomy is earned on evidence rather than granted on trust.
Every day will be different, but you can expect it to include:
Build security into the path to production. You turn our baseline standards into checks that run on every change: policy as code, review agents that check each change against the baseline and a team's own standards, and gates matched to the risk of what's changing. Staying on the path should meet the security requirements without any extra work.
Secure the agents we and our colleagues build. You threat-model agent systems end to end, from prompt injection and tool misuse to data leaking through context and retrieval, and design the controls that stop it: scoped tool permissions, permission-aware retrieval, sandboxing, and human approval where the risk calls for it.
Give every agent an identity. You build the identity and access layer for agents. Each one has its own identity, a named owner, least-privilege access that is reviewed and can be revoked at once, and a full record of what it did.
Build detection and response with agents. You write detections and automated responses on our SIEM and SOAR, and build agents that triage alerts, gather context and carry out bounded, reversible responses, so people spend their time on the cases that need judgement.
Secure the workplace, worldwide. You apply the same approach to the devices, workplace tools and accounts colleagues use every day: automated controls rather than manual checklists, and access that is granted and removed as people join, move and leave.
Prove it holds with security evals. You build eval pipelines and red-team exercises that show whether our agents and controls stand up to attack, and produce compliance evidence as a by-product of the work rather than as screenshots for an auditor.
Influence technical direction. You have opinions about how things should be built, and you back them up with evidence. You contribute to architectural decisions, push back when the team is heading in the wrong direction, and propose better approaches.
Security Engineering: You have built security into systems rather than reviewed it after the fact, across some mix of application and cloud security, identity and access, and detection and response. You understand how AI systems get attacked, through prompt injection, tool misuse and data leaking through context, and you design controls that make the secure option the easy one for the people using them
Production AI Agent Engineering: You have shipped AI systems that serve real users at scale. You understand the engineering challenges that make agent systems different from conventional software, including context management, model selection and routing, cost engineering, tool use and agent augmentation, and evaluation
Product mindset: Your natural instinct is to identify the real user pain, understand their requirements and build pragmatic, high-impact solutions. You test, learn and iterate fast, and are comfortable executing under high ambiguity
Full-Stack Delivery: You work across the stack: LLM integration, backend services, data pipelines, and enough frontend to ship end to end
Communication: You can explain technical concepts clearly to both engineers and business colleagues. You proactively engage the people around your work, and bring them along with you on the journey
Fluent in AI-assisted development and can mentor others to do it well
Benefits
Time off: 27 days holiday (including 3 winter shutdown days) and bank holidays, plus 5 extra days off (1 life event, 2 volunteer days and 2 summer shutdown days).
Our Commitment to Diversity, Equity and Inclusion
Standard company text repeated across Multiverse IO's postings is omitted here.