Key Responsibilities
• Contextual Business Translation Translating high-level industry problems and domain-specific business needs into concrete data science, prompt engineering, and LLM-centric requirements.
• AI Value Maximization & Backlog Engineering Prioritizing product backlogs based on algorithmic complexity, data availability, and business impact to maximize the ROI of AI features.
• AI Requirements & Synthetic Stories Discovering and managing requirements specifically tailored for machine learning and cognitive computing systems. Authoring synthetic user stories, acceptance criteria, and edge cases for systems where the output is probabilistic rather than deterministic.
• Process Re-imagining & Agentic Workflows Re-engineering current-state process flows to embed agentic workflows and automated decisioning. Mapping out human-in-the-loop feedback mechanisms for continuous model optimization cycles.
• Governance & Probabilistic SLAs Defining product success and performance guardrails for non-deterministic AI outcomes, such as accuracy thresholds, latency, and hallucination rates. Translating organizational ESG, compliance, and responsible AI guardrails into specific system requirements.
Primary Skills
• Generative AI & LLM Familiarity Deep understanding of the capabilities, constraints, and architecture of large language models and cognitive technologies, including commercial and open-source options, tokenomics, and cost optimization levers.
• Agentic System Architecture Strategy Understanding of how autonomous agents operate, with the ability to architect strategies around when to deploy point solutions versus autonomous, multi-agent automated systems.
• Prompt Engineering & Model Steering Ability to evaluate and refine system prompts, guardrails, and context windows to ensure alignment with product brand guidelines and craft advanced semantic structures for data synthesis.
