AI Platform/Agent Engineer (Claude)

Responsibilities: Develop AI agents that investigate security alerts, gather evidence, and produce structured recommendations with confidence scores. Integrate Claude models with Splunk and other enrichment sources using secure, read-only access. Build evaluation frameworks using historical alert and analyst disposition data. Optimize prompts, workflows, and model performance for accuracy, cost, and latency. Implement secure cloud deployments, CI/CD pipelines, observability, and AI security controls. Conduct shadow-mode validation and support knowledge transfer to internal teams.

Required Qualifications: Hands-on experience with Claude APIs, agent frameworks, MCP, and AI workflow orchestration. Strong Python development skills, including asynchronous and event-driven architectures. Experience building production AI applications with monitoring, logging, and cost management. Expertise in cloud deployment, infrastructure automation, secrets management, and API integrations. Experience evaluating AI systems using accuracy, false-negative analysis, regression testing, and model governance.

Preferred Qualifications: Experience integrating AI solutions with Splunk or similar security platforms. Knowledge of prompt injection mitigation and AI application security best practices. Experience supporting human-in-the-loop workflows and operational AI systems.

Success Metrics: Deploy a production-ready AI alert triage solution operating in shadow mode. Achieve defined accuracy targets, including low false-negative rates and high analyst agreement. Deliver complete documentation, source code, testing frameworks, and operational runbooks to support long-term ownership and maintenance.

Rate: $75-$90/hr (Dependent on level of experience and fit)