AI Agent Architect - LLM Systems

Cyngular is building an AI-native cloud security platform focused on investigation, threat hunting, and autonomous analysis across multi-cloud environments. We are looking for a senior AI Agent Architect to help design and refine the next generation of our AI-driven investigation and reasoning engines.

This is a strategic advisory role, working closely with the engineering team.

What You’ll Do

• Architect autonomous AI agents for investigation and cloud threat analysis
• Design multi-step reasoning pipelines for structured decision-making
• Build and refine RAG architectures with measurable evaluation frameworks
• Define memory strategies (short-term, long-term, vector-based, graph-based)
• Optimize LLM cost, latency, and token efficiency at scale
• Identify and mitigate failure modes in AI systems (hallucinations, loops, tool misuse)
• Establish evaluation and benchmarking methodology for AI agents
• Guide internal engineers in production-grade AI system design

We’re Looking For Someone Who

• Has built autonomous AI agents in production environments
• Designed complex reasoning workflows beyond simple prompt chains
• Implemented RAG systems with proper retrieval tuning and evaluation
• Optimized large-scale LLM usage with cost-awareness and performance tradeoffs
• Deeply understands AI system failure patterns and guardrail design

Nice to Have

• Experience with multi-agent coordination frameworks
• Background in cloud-native architectures
• Strong distributed systems experience
• Familiarity with security, SOC workflows, or threat hunting domains

Engagement Structure

• Advisory role on a flexible hour-based retainer
• High-impact architecture and design involvement
• Direct collaboration with executive and engineering leadership
• Opportunity to shape AI-driven cloud security at scale

If you have built real AI agents that operate in production and understand the engineering tradeoffs behind them, we’d like to talk.