About this role.
Maze is hiring a senior Machine Learning Engineer to lead the productionization of LLM- and agent-based cybersecurity capabilities. The role owns evaluation frameworks, ML experimentation-to-production pipelines, monitoring, and integration of ML features into the customer product. It requires at least six years of production ML experience, strong Python and deep-learning foundations, and the ability to work directly with the CTO while operating independently. The engineer will improve agent performance, build scalable MLOps foundations, and mentor junior engineers through technical leadership.
Role DNA
A quick view of the complexity, pace, ownership and collaboration implied by the job description.
Job Complexity
5/5Pace & Pressure
5/5Autonomy Level
5/5Communication Load
4/5Salary analysis
Estimated compensation compared with the broader UK market for similar roles.
Core skills
Skills and capabilities most closely associated with this opportunity.
Sample interview questions
I would explain the initial customer problem and success metrics, then cover dataset and model choices, offline evaluation, deployment architecture, monitoring, rollback controls, and iteration process. I would quantify production outcomes such as latency, reliability, adoption, model-quality improvement, or business impact.
I would define task-specific success criteria, including correctness, safety, groundedness, latency, cost, and analyst usefulness. I would combine curated adversarial test sets, production traces, automated scoring where reliable, human review for high-risk cases, versioned benchmarks, and dashboards that compare agent versions over time.
I would separate fast prototyping from controlled production release paths. Feature flags, reproducible experiments, versioned prompts and models, automated regression evaluations, observability, staged rollouts, and clear rollback procedures allow teams to move quickly without exposing customers to uncontrolled quality or security risks.
I would first segment failures using traces and evaluation data to identify whether the issue is retrieval, prompting, tool use, model selection, workflow design, or ambiguous task definition. I would test focused hypotheses, measure improvements against a fixed benchmark and live guardrail metrics, then deploy the smallest validated change through a staged rollout.
I translate technical choices into user outcomes, risks, and measurable trade-offs such as accuracy, response time, cost, coverage, and failure modes. I use concrete examples, establish agreed success metrics early, and clearly state what the system can and cannot reliably do so stakeholders can make informed product decisions.
Summary of the Role:
As ML Engineer at Maze, you’ll be the technical leader driving our machine learning infrastructure from experimentation to production, ensuring our AI-powered cybersecurity solutions deliver measurable impact for customers worldwide. This is a unique opportunity to join as one of the early engineering team members of a well-funded startup building breakthrough applications of LLMs and AI agents in cybersecurity.
You’ll take full ownership of evaluation frameworks, production ML pipelines, and cross-team ML integration, working closely with our CTO and product teams to transform cutting-edge AI research into robust, scalable solutions that solve real security challenges. Your success will be measured by agent performance improvements and product innovation impact, not just technical metrics. This role is perfect for a hands-on ML engineer who has scaled production ML systems across multiple companies, thinks like a product builder, and wants to drive the actual productionization of LLMs and ML to solve significant pain points.
Your Contributions to Our Journey:
Build Production-Grade Evaluation Systems: Design and implement comprehensive evaluation frameworks that measure agent performance, track improvements over time, and ensure our AI systems deliver consistent value to customers
Drive Experimentation-to-Production Pipeline: Own the entire ML lifecycle from prototype to production, building scalable systems that enable rapid iteration while maintaining reliability and performance in customer environments
Enable Cross-Team ML Integration: Work closely with product teams to seamlessly integrate ML capabilities into customer-facing features, ensuring technical excellence translates into user value and product differentiation
Optimize AI Agent Performance: Continuously improve our AI agents through systematic experimentation, prompt engineering, and architectural enhancements, measuring success through customer impact and system performance
Scale ML Infrastructure: Build the foundational ML systems, monitoring, and tooling that will support our growth from startup to scale, ensuring we can deploy new capabilities quickly without compromising quality
Partner with Engineering Leadership: Collaborate directly with our CTO through regular check-ins and strategic alignment while operating with high autonomy and self-direction in day-to-day execution
Mentor Through Excellence: Provide natural mentorship to junior ML engineers through code reviews, technical guidance, and sharing practical experience from building production ML systems
What You Need to Be Successful:
Proven Production ML Experience: 6+ years building and scaling machine learning systems in production environments, with hands-on experience moving from experimentation to customer-facing deployments
Deep Neural Networks Foundation: Strong background in classical neural networks and deep learning fundamentals before specializing in modern LLMs and transformer architectures – you understand the foundations, not just the latest tools
Product-Focused ML Mindset: Experience building ML systems that solve real business problems, with a track record of integrating classification, prediction, or recommendation systems into actual products customers use
Multi-Company Perspective: Experience across multiple organizations (scale-ups, startups, or combination), giving you practical knowledge of what tools to build vs buy and how to avoid over-engineering
Technical Versatility: Strong Python skills with flexibility across ML frameworks and tools – comfortable adapting to our stack including LangChain, evaluation frameworks, and workflow orchestration tools like Temporal
Self-Directed Leadership: Ability to operate autonomously while maintaining close alignment with leadership, comfortable with frequent check-ins but capable of driving projects independently
Cross-Functional Collaboration: Experience working closely with product teams and potentially customers, translating technical capabilities into business value and user experiences
Nice to Haves:
Experience with AI agents, LLMs, or modern generative AI applications
Cybersecurity domain knowledge or experience applying ML to security challenges
Background at ML-first companies or organizations where ML was core to the product
Experience with modern MLOps practices and cloud-based ML infrastructure
Track record of optimizing model performance and controlling AI system costs
Why Join Us:
Real-World AI Impact: Drive the actual productionization of LLMs and machine learning to solve significant cybersecurity pain points – your work will directly protect organizations from real threats, not just optimize internal metrics
Technical Leadership Opportunity: Work directly with our CTO on cutting-edge ML infrastructure while having the autonomy to shape technical decisions and build systems that scale with our hypergrowth
Expert Team Partnership: Join a team of hands-on leaders with experience in Big Tech and Scale-ups, including leadership team members who have been part of multiple acquisitions and an IPO
Build the AI-Native Future: Shape how generative AI transforms cybersecurity from the ground up, establishing ML practices and technical standards that will define the industry
Multiple Growth Pathways: Clear opportunities to grow into Head of ML Engineering, become a domain technical lead, move into customer-facing technical roles, or excel as a senior individual contributor – the choice is yours based on your interests and our needs
Breakthrough Technology: Work at the intersection of generative AI and cybersecurity, building solutions that leverage the latest advances in LLMs and AI agents to solve some of the most pressing challenges security teams face today
Upload your resume now
- To unlock remote work opportunities and be discovered by global employers.
This job listing has been manually reviewed by the Jobicy Trust & Safety Team for compliance with our posting guidelines, including verification of the company's legitimacy, accuracy of job details, clarity of remote work policy, and absence of misleading or fraudulent content.
