Role overview
A Prompt Engineer role focused on designing, evaluating, and systematically optimizing prompts for large language models used in client-facing AI systems. The position sits at the intersection of linguistics, software engineering, and applied ML, with responsibility for closing the gap between working demos and production-grade LLM behavior. It is a fully remote, US-based, full-time opportunity with a salary range of $130,000–$190,000.
Responsibilities
- Analyze target use cases to design prompting strategies using few-shot examples, chain-of-thought reasoning, structured outputs, self-consistency, and multi-step workflows with verification
- Build evaluation frameworks that measure accuracy, consistency, latency, cost, and safety across prompt variants
- Implement A/B testing infrastructure, golden datasets for regression testing, and LLM-as-judge evaluation chains
- Develop reusable prompting patterns and documentation that can be adapted across client use cases
- Create tooling and processes that make prompt development more scientific and less reliant on guesswork
- Iterate empirically, using measured results to drive prompt and system improvements
Requirements
- Hands-on experience prompting and evaluating large language models in production or near-production settings
- Strong written communication and ability to reason carefully about language, instruction following, and edge cases
- Familiarity with modern prompting techniques and their tradeoffs across different model families
- Experience designing experiments, running A/B tests, and interpreting evaluation results
- Ability to build light tooling or scripts that support prompt iteration and regression testing
- Comfort working across multiple client engagements and use cases simultaneously
Nice to have
- Background in linguistics, cognitive science, or human-computer interaction
- Public speaking or conference experience on applied LLM work
Benefits and work setup
- Fully remote within the US, full-time
- Salary range of $130,000–$190,000
- MacBook Pro, external display, and peripherals of choice
- Access to AI tools and subscriptions such as GPT-4, Claude, and GitHub Copilot
- 401(k) with company match, 12 weeks paid parental leave, and a $1,000 home office stipend
- Quarterly team offsites in interesting locations
- Support to attend or speak at AI conferences