Vacancy details
This is a private vacancy. Company details stay confidential until InVault confirms a serious candidate fit.
- Status
- Active
- Date posted
- 2026-10-12
- Open until
- 2027-01-10
- Employment type
- Full-time
- Source-listed languages
- Not specified
- Language requirement
- No mandatory language inferred
- Work arrangement
- Hybrid
- Physical workplaces
- Sofia, Bulgaria
- Applicant eligibility
- Applicant geography not specified
Overview
This security role combines established infrastructure controls with the newer risks of model and agent deployments. The scope includes locally hosted and cloud AI systems, the services around them and the identities that connect tools with data. Work is mainly directed at assessing controls, prioritising improvements and guiding teams through implementation. Endpoint and email-security technologies remain part of the wider responsibility. In a regulated operating environment, recommendations need clear evidence and documentation so that engineering, security operations and compliance colleagues can understand the remaining exposure and follow remediation. Repeatable controls and audit-ready records are essential outputs.
Requirements
- Relevant certifications or verifiable security/IT experience.
- Infrastructure security, EDR and cloud/local LLM knowledge.
- Understanding LLM threats and AI-security frameworks.
- Guardrail assessment and regulated-environment experience.
Responsibilities
- Assess LLM deployments, gateways, vector stores and agent/MCP components.
- Improve endpoint hardening, EDR and related security technologies.
- Review prompt-injection controls, permissions, secrets and AI supply chains.
- Recommend logging and detection coverage for AI workloads.
- Support security assessments, audits and repeatable control guidance.
Conditions
- Hybrid arrangement in Sofia, Bulgaria.
- Full-time employment.