Senior Product Engineer, Out-of-Band Policy Engine, AI Platform

Redpanda Data

. Take the out-of-band policy engine from benchmarked prototype to production .

Posted 9/29/2026full-timeWarsaw • United StatesSenior💰 $205,000 - $230,000 per yearWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in policy-as-code systems, particularly Cedar and Open Policy Agent, with a strong focus on security-critical production environments. Proven ability to collaborate with distributed teams and communicate effectively while managing technical ownership and adversarial testing.

Highest-signal resume keywords

Policy-As-Code SystemsCedarAdversarial TestingSecurity-Critical Production ProxyAWS, GCP, or Azure

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

Software DevelopmentAuthorization FrameworksAI Agent SystemsDistributed-Systems DesignAudit/Compliance Logging PipelinesRed-Team Evaluation HarnessesLLM Guardrail SystemsSQL AnalyticsStreaming Data ProcessingTechnical Ownership

Soft Skills

Strong Verbal CommunicationStrong Written CommunicationCollaboration

Tools & Technologies

GitHubCedar-Based Enforcement ProxyAI GatewayGovernance Platform

Industry Keywords

Security-Conscious IndustriesCompliancePolicy DecisionsData ExposureSemantic Gating

Tech Stack

Tools & technologies

AWSAzureGoogle Cloud PlatformSQL

About the role

Key responsibilities & impact

Take the out-of-band policy engine from benchmarked prototype to production
Wire the Cedar-based enforcement proxy into the AI gateway as the guardrail path for governed tool calls
Build the two-tier policy composition model with per-tool-server data-owner ceilings and per-agent narrowing
Build the authoring experience for security and compliance teams to write, test, and validate policies
Extend authorable policy operators across authorization, data exposure, and semantic gating
Implement deferred approval routing for human review
Record every enforcement decision as an audited, attributable, replayable record
Maintain and extend the adversarial certification harness with a multi-turn red-team adversary, blinded judge, and deterministic leak metrics
Work directly with design-partner customers in security-conscious industries
Raise difficult or systemic challenges and impediments to the manager
Discuss strategic topics with peers to shape the product roadmap and team practices
Track progress, assess risks, and communicate contingency and mitigation plans
Develop Redpanda's governance platform for agent-data interaction, combining streaming, SQL analytics, intelligent connectivity, and AI-agent governance

Requirements

What you’ll need

5+ years of experience in software development, including building systems that enforce authorization or policy decisions in a live request path
Experience with policy-as-code systems such as Cedar, Open Policy Agent (Rego), XACML, or comparable authorization frameworks
Solid understanding of AI agent and LLM system failure modes, including prompt injection, tool misuse, and data exfiltration through model output
Experience with adversarial testing and designing for attackers
Experience collaborating with a globally distributed engineering team and working with GitHub in the open
Strong verbal and written communication skills
Demonstrated technical ownership
Direct experience with Cedar or deep experience with a comparable policy-as-code language is highlighted as applicable
Experience building or operating a security-critical production proxy or gateway is highlighted as applicable
Experience designing or running red-team/adversarial evaluation harnesses, including LLM-as-judge setups, is highlighted as applicable
Experience with LLM guardrail, content-safety, or prompt-injection defense systems is highlighted as applicable
Experience with audit/compliance logging pipelines is highlighted as applicable
Application form asks about 8+ years of professional software engineering experience
Application form asks about distributed-systems design and architecture expertise
Application form asks about hands-on agentic AI infrastructure experience
Application form asks about regular use of AI agents or subagents
Application form asks about strong knowledge of AWS, GCP, or Azure
Warsaw applicants must be able and willing to work onsite at least 3 days per week

Benefits

Comp & perks

Latest AI tools
Budget to use AI tools
People-first organization
Culture based on trust, transparency, communication, and kindness
Remote work for the US/Canada Eastern Region role
Hybrid work option for the Warsaw, Poland location