Security Analytics Architect
OpenKyber LLC · Atlanta, GA, United States ·
COMPENSATION
€•• – ••k
Listed on WeAreDevelopers
- Onsite in Fort Worth, TX
- Contract position
- AI & LLM Security
- LangChain & LangGraph
- AWS & Azure
We are looking for a hands-on technical architect to lead security controls for AI/ML systems. You will serve as the architecture authority, guiding engineers and driving secure patterns for data, training, and deployment while mitigating threats like prompt injection and model poisoning.
What you'll do
- Define RBAC and least-privilege models for AI agents and non-human identities
- Design runtime environments with restricted permissions to prevent unauthorized access
- Implement defenses against adversarial attacks, prompt injections, and data leakage
- Architect logging standards to ensure traceability of agent reasoning and API calls
- Monitor production models for behavioral drift, anomalies, and poisoning attacks
- Map AI-specific controls to standards like NIST AI RMF, OWASP, and GDPR
- Conduct AI/ML threat modeling and translate findings into remediation backlogs
- Mentor engineers on secure design implementations without line management duties
What you'll need
- 7+ years in cybersecurity architecture with large-scale LLM experience
- 5+ years hands-on with agent frameworks like LangChain, LangGraph, or AutoGen
- 3-5 years familiarity with model risk management and AI security standards
- Experience securing multi-agent workflows and MLOps platforms
- Knowledge of data residency, privacy laws, and SOC 2 Type II compliance
- Ability to disclose model providers and training data practices clearly
Stack
- LangGraph
- LangChain
- AutoGen
- AWS
- Azure
- M365 Copilot Studio
- Entra Agent ID
How you'll work
- Onsite in Fort Worth, Texas
- Contract employment type