Design, build, and deploy a AI-driven security alert triage and investigation platforms leveraging the Claude ecosystem and Splunk.
Key responsibilities included:
- Building multi-step AI agents using Claude APIs, Agent SDK, MCP, and prompt engineering techniques.
- Integrating Splunk and enrichment sources to automate alert investigation and evidence collection.
- Developing evaluation frameworks using historical security alerts and analyst dispositions to measure accuracy and reduce false negatives.
- Executing shadow-mode deployments, regression testing, and continuous prompt/model optimization.
- Implementing production-grade solutions with Python, cloud infrastructure, CI/CD pipelines, secrets management, observability, and API resilience.
- Establishing monitoring for cost, latency, error rates, agent tracing, and operational performance.
- Delivering secure AI systems with safeguards against prompt injection, governance controls, documentation, and knowledge transfer.
Skills: Claude Platform, Anthropic APIs, Agentic AI, MCP (Model Context Protocol), Splunk, Python, Prompt Engineering, AI Evaluation, Cloud Architecture, CI/CD, Observability, Security Automation, Production AI Systems, LLM Engineering.
Pay around $80-100/hr
AI Engineer (Claude) in chicago at Unknown Company
This position is listed as full time and onsite.