Role Summary: Designs and governs multi-agent AI architecture for vulnerability triage and automated remediation using Vertex AI GPT Claude.

Required Skills:

• Vertex AI, GCP AI services, prompt engineering LLM (GPT, Claude, Gemini), Lang Chain Semantic
• Kernel Vector DB (Pinecone, FAISS, Vertex embeddings)
• Python, ML pipelines AI Agentic Expectations Deep expertise in
• Multi-agent collaboration LLM judge scoring models Autonomous decision making fallback logic
• Design secure, explainable AI outputs

Key Responsibilities:

• Design multi-agent architecture (triage agents, remediation agents, judge agent)
• Build LLM orchestration (Vertex AI Lang Chain Agent frameworks)
• Define RAG pipelines using secure coding knowledge packs
• Establish model evaluation framework (accuracy, false positives, latency)
• Design agent memory (vector DB, knowledge graph)
• Implement prompt engineering, chaining, tool usage delegation
• Define model versioning, evaluation harness, and governance