- Design, test, and refine prompts to maximize output quality, reliability, and efficiency across client's use cases.
- Collaborate closely with the AI/engineering team to understand system architecture and embed prompt strategies into workflows.
- Run structured experiments (prompt variants, chaining, few-shot/zero-shot techniques) and document what measurably improves outcomes.
- Identify and mitigate common failure modes – hallucination, inconsistency, prompt injection risk – within the system's use cases.
- Translate business/functional requirements into effective prompt design patterns and reusable templates.
- Provide clear, concise reporting on findings and impact to support a go/no-go decision on further investment.
Required Qualifications
- Certified Prompt Engineer (formal certification required – not a general AI/software engineering profile).
- Hands-on experience designing, testing, and optimizing prompts for LLM-based applications, preferably through project, pilot, or production environments.
- Good understanding of LLM behavior, context windows, tokenization, and common prompting techniques such as few-shot prompting, chain-of-thought, and retrieval-augmented prompting.
- Ability to work independently and deliver measurable results within a short, defined engagement window.
- Solid documentation and communication skills to support experimental reporting.
Preferred / Nice-to-Have
Prior experience in experimental or innovation-lab style engagement with tight timelines.
Familiarity with enterprise AI platforms and integration patterns.
Exposure to agentic AI / multi-step reasoning systems.