Anthropic Design Developer

Anthropic Design Developer | Location: Chicago, IL or Dallas, TX| Contract | Hourly Pay Range: $90-$105 We are seeking an experienced AI Design/Development Engineer to lead the creation of an AI-driven alert triage and investigation system on the Claude Platform, integrated with Splunk. This long-term contract role involves designing, building, and deploying scalable AI workflows, ensuring production readiness, and delivering comprehensive documentation and knowledge transfer.

Key Responsibilities

• Architect and develop agent workflows utilizing the Claude Platform, Messages API, and MCP for alert review, evidence gathering, and verdict generation.
• Implement secure deployment practices, including secrets management, logging, tracing, and safeguards against prompt injection.
• Build an evaluation framework using historical alert data and analyst dispositions to measure system accuracy, false-negative rates, and agreement with analyst decisions.
• Optimize system performance through cost and latency management techniques such as prompt caching, model tiering, batching, and token budgeting.
• Collaborate with internal security teams to incorporate domain expertise, triage criteria, escalation rules, and labeled datasets.
• Conduct shadow-mode testing, monitor system metrics, and iterate until predefined accuracy thresholds are achieved.
• Deliver all source code, prompts, tools, datasets, documentation, and runbooks, ensuring the client can independently maintain and upgrade the system.

Required Qualifications

• Hands-on experience with the Claude Platform, Messages API, and MCP for orchestrating multi-step agent workflows.
• Proficiency in Python, including async and queue-based architectures, with a focus on idempotent processing.
• Strong knowledge of cloud deployment, CI/CD pipelines, infrastructure as code, secrets management, and API rate limiting.
• Experience with observability tools for tracing, logging, and dashboard creation to monitor latency, cost, and error rates.
• Demonstrated ability to build and evaluate datasets from historical alerts and analyst feedback, with a focus on accuracy and false-negative control.
• Proven experience in system testing, regression testing, and managing model version upgrades.
• Ability to deploy in shadow mode, compare AI and analyst results, and ensure system robustness before production.

Preferred Qualifications

• Familiarity with the latest Anthropic documentation and model features.
• Experience with prompt engineering for structured outputs, including verdicts, confidence levels, and evidence citations.
• Knowledge of security best practices for prompt injection safeguards.
• Prior experience in security or SOC environments is not required but beneficial.

About Delphi-US - Peacemakers In The Talent War Delphi-US is a national IT Services firm based in Newport, Rhode Island. Delphi attracts our nation's best and brightest Technical & Professional Talent and positions our nation's finest employers to further their competitive advantages. Delphi's core technical focus includes Data Intelligence Engineering & Analytics, AI/ML Operational Excellence, DevOps & Cloud Computing, Health IT & Life Sciences, and Energy Management/Power Systems; serving domains from sensitive national energy sectors to critical financial services, biomed, life sciences & pharma, major government & municipal institutions, and much of the USA's Fortune 500 corporations. Delphi Associates are deployed on critical IT Projects where their consultative business acumen, comprehensive technical expertise, and committed service mission drives our client's success. Our proprietary skill-based and cultural matching process results in higher qualified project teams, custom curated for our client's specific technical & functional goals. We are supremely experienced, friendly, professional, and ready to advocate on your behalf. Delphi has an indelible understanding of employer expectations, a track record of project success and we deliver lasting results.

6398