AI Engineer

Objective To architect, orchestrate, and maintain a fully autonomous, hands-off work environment. Your primary mission is to build the AI infrastructure and agent workflows necessary to execute and deliver Rendy's projects from end to end with zero manual operational overhead.

Core Responsibilities

• System Protocol Engineering: Design, write, and continuously optimize core agent instructions and configuration files, specifically mastering skills.md, AGENT.md, and claude.md to ensure agents behave exactly as required.
• Workflow Automation: Develop hands-off execution pipelines where AI agents can independently plan, execute, debug, and review tasks required to get Rendy's projects done.
• Prompt Architecture: Craft highly advanced, context-rich prompts that guide LLMs to produce production-grade code, architectures, and deliverables on the first pass.
• Agent Orchestration: Set up environments where multiple AI tools and agents interact seamlessly, delegating tasks and communicating effectively without human intervention.
• Quality Assurance & Auditing: Implement automated checks and balances within the AI workflow to catch hallucinations, logical errors, or deviations from Rendy’s project requirements.

Required Knowledge & Skills

• Deep Model Expertise: Comprehensive understanding of state-of-the-art LLMs (Claude, Gemini, GPT-4, etc.), including their context windows, rate limits, strengths, and specific modalities.
• Advanced Prompt Engineering: Ability to structure complex, multi-step prompts, utilize few-shot learning, and design dynamic system instructions that enforce strict formatting and operational constraints.
• Tooling & Ecosystems: Familiarity with AI agent frameworks, vector databases, RAG implementations, and API integrations.
• Critical Thinking: Exceptional ability to anticipate edge cases, dissect complex project requirements into atomic AI-executable tasks, and troubleshoot recursive agent loops.
• Extreme Attention to Detail: A meticulous eye for structural consistency in markdown files, syntax validation, and identifying subtle flaws in AI-generated output before they compound in an automated pipeline.