What are the responsibilities and job description for the Lead AI Security & Architecture Engineer- PAM position at Cloud Destinations LLC?
PAM ROLE
Locations: Location for this role is New York, Montreal or Alpharetta, GA. 3x a week onsite
Key Responsibilities:
- Lead hands-on development of AI enabled and LLM based applications, including agentic and automation?driven systems.
- Design and implement agent orchestration architectures, including task decomposition, multi?agent coordination, tool/function invocation, state and memory management, and policy-aware execution flows.
- Engineer robust LLM interaction layers, including prompt design, grounding strategies (e.g., RAG), tool integration, feedback loops, and evaluation mechanisms.
- Own end-to-end AI system architecture, spanning APIs, services, data pipelines, model serving, and observability.
- Ensure AI solutions operate effectively across cloud-native and hybrid environments, with attention to scalability, latency, and reliability.
- Embed security, compliance, and governance-by-design, including access controls, logging, traceability, explainability, and human?in?the?loop safeguards.
- Provide technical leadership through architecture ownership, hands-on coding, design reviews, and mentorship of AI engineers.
Required Skills:
- Proven experience as a hands-on AI or platform engineer with leadership responsibility for production systems.
- Deep expertise in LLMs, including model selection, prompt engineering, grounding techniques, evaluation, and mitigation of hallucination and drift.
- Strong experience with agent frameworks and orchestration patterns, including multi?agent systems, tool?using agents, and agent lifecycle management.
- Solid background in cloud?native architecture, APIs, distributed systems, and modern MLOps/LLMOps practices.
- Ability to translate business, risk, and regulatory requirements into concrete technical designs and implementations.
- Experience translating advanced AI (LLMs, agentic workflows, orchestration) into secure, governed, and auditable capabilities that are production?ready for large, regulated enterprises.
- Experience with real-world deployments scenarios ensuring AI solutions work reliably across hybrid environments, integrate with enterprise platforms, and deliver measurable business outcomes.
Preferred Qualifications:
- Experience implementing AI control frameworks (e.g., model controls, guardrails, evaluation, and auditability) aligned to NIST, ISO, or sector regulators.
- Knowledge of identity, access, and authorization models for agents and non?human identities, including least?privilege and JIT patterns.
- Familiarity of frameworks such as OWASP Top 10 for Agentic and LLM Applications, MITRE ATLS and NIST AI RMF.