Staff AI Security Architect

Description

The company runs a customer success platform that covers onboarding, outcomes and advocacy, and is used by more than 2,000 companies through its applications and AI agents.

This is a full-time Staff AI Security Architect role on the Security team, reporting to the Director of Security. It is hybrid, based in Wrocław, Poland.

You will help secure the company's AI-powered products, internal AI tools and customer trust, working closely with Engineering, Product, Data, IT, Legal and GRC in a fast-moving AI environment.

Responsibilities

• You will lead threat modeling and security design reviews for LLM-based features and agentic workflows.
• You will set reference architectures and guardrails for building and adopting AI, and assess risks from third-party models and AI vendors.
• You will turn changing AI risk guidance into practical engineering standards and advise leaders on tradeoffs.
• You will explain complex AI risks to technical and executive audiences and drive decisions without formal authority.
• You will work hands-on as a senior security practitioner.
• Occasional travel of up to 10-25% may be required for team meetings, training or company events.
• The list is not exhaustive, and the scope may change as the team and business needs change.

Requirements

• You have 8+ years in security architecture, application security or security engineering, including hands-on work securing AI/ML or LLM-powered applications.
• You understand AI-specific threats and their mitigations, such as prompt injection, data leakage, model and supply chain risk, insecure tool or plugin use and excessive agent autonomy, including the OWASP Top 10 for LLM Applications and MITRE ATLAS.
• You have led threat modeling and security design reviews for complex systems and turned the findings into clear, actionable architecture requirements.
• You have strong knowledge of cloud-native architecture on AWS, Azure or GCP, covering identity and access management, secrets management, data protection and API security for AI workloads.
• You have designed guardrails and controls for AI systems, such as input and output filtering, access control for retrieval and agent tools, logging, monitoring and secure handling of sensitive customer data in AI pipelines.
• You have working knowledge of AI governance and risk frameworks such as NIST AI RMF and ISO/IEC 42001, and can turn them into engineering requirements.
• You can partner with Engineering and Product leaders, influence without authority and explain technical risk to executives and non-technical audiences.
• Nice to have: hands-on experience with agentic systems, MCP, or ML engineering, including building or deploying LLM applications, RAG pipelines, custom MCP servers or agentic systems.
• Nice to have: experience with AI red teaming or adversarial testing of LLM applications and agentic systems.
• Nice to have: experience securing MLOps and model-serving pipelines, including model provenance and training data protection.
• Nice to have: experience supporting enterprise customer security and AI governance reviews for a B2B SaaS product.
• Nice to have: security certifications such as CISSP, CCSP or GIAC.

Conditions

Full-time, hybrid role based in Wrocław, Poland.

Reports to the Director of Security.

If you apply through an Employer of Record (EOR) or as a contractor, employment terms and benefits are managed by the EOR or may not apply to non-EOR contractors.

The company is an equal opportunity employer and welcomes candidates from all backgrounds.

Benefits

• Growth opportunities, including mentoring and career development.
• The company describes a culture of stability, trust and support, and a strong sense of community.