defining end-to-end architecture, version control, CI/CD pipelines, and deployment playbooks for distributed teams.
Design complex, high-impact prompt systems including multi-agent and chain-of-thought architectures for resume builder flows across global product lines.
Partner with Engineering, Product, and Data to align prompt strategies with model selection, API cost reduction, performance budgets, and roadmaps.
Architect evaluation frameworks and safety guardrails to benchmark LLM outputs globally, minimizing hallucination, bias, and AI-sounding content.
Act as the cross-functional prompt design resource; translating requirements from Content, Engineering, Data, Product, and portal teams into production-grade implementations.
Spearhead R&D into emerging AI paradigms, evaluate new models and tooling, and provide clear adoption recommendations.
Mentor the prompt management team and contribute to building the function
structures, processes, documentation, and quality standards.
Requirements
8+ years of professional experience with AI/ML systems, language technologies, or software product engineering.
4-5 years in prompt design or LLM product development, with a track record of shipping consumer-facing features at scale.
Mastery of advanced prompt design paradigms: RAG, multi-agent architectures, chain-of-thought, context window management, and model behaviour trade-offs.
Hands-on experience across LLM providers (OpenAI, Anthropic, Google) with clear understanding of behavioural differences and trade-offs.
Experience designing and maintaining LLM evaluation pipelines (LLM-as-judge, human eval, automated metrics) and prompt versioning workflows.
Strong instinct for naturalness, recognising and eliminating AI-sounding, repetitive, or generic output.
Demonstrated ability to influence cross-functional roadmaps and mentor prompt designers or AI content teams.