Applied AI Engineer (.NET/C#)

Key Activities

• Design prompts, context strategies, and LLM integrations for product features involving business-critical prices, parts, or stock information.
• Integrate AI capabilities into established C# .NET codebases through clean service boundaries and sensible abstractions.
• Prototype in days, harden in weeks, and make pragmatic decisions about delivery trade-offs.
• Build evaluations alongside features, test real business cases, and use measured results to guide decisions.
• Work with existing engineering teams, product leaders, and subject matter experts to understand requirements and customer needs.
• Choose models, tools, and providers pragmatically based on quality, cost, and latency.

Systems, Tools and Technology

• C# .NET is the primary development language.
• The role uses large language model APIs across multiple providers.
• The role involves GitHub Copilot, Cursor, or equivalent AI coding tools.
• Prompt engineering, context design, retrieval-augmented generation, vector search, and embeddings are relevant technologies.
• Evaluation frameworks and automated quality pipelines support AI output quality.

Required Qualifications

• Solid production experience with C# .NET, including work in established codebases and safe delivery of changes.
• Hands-on experience building with large language models, including prompt design, context engineering, and structured outputs.
• A track record of shipping working software from idea to delivery in weeks rather than quarters.
• Experience evaluating LLM outputs in a structured way and using results to improve quality.
• Daily fluency with AI coding tools such as GitHub Copilot, Cursor, or equivalents.

Preferred Qualifications

• Production experience with retrieval-augmented generation, agentic workflows, tool use, vector search, or embeddings is preferred.
• Experience with LLM APIs from more than one provider is preferred.
• Exposure to Klipboard sectors and experience modernizing long-lived systems are preferred.
• Familiarity with evaluation frameworks, test datasets, or automated AI quality pipelines is preferred.

Success Measures

• Ship AI capability into at least one established product with evaluation behind it within the first six months.
• Deliver a feature from idea to customers in weeks that holds up in production.
• Use evaluation results to change decisions, including stopping work that is not good enough to ship.
• Help surrounding engineers learn techniques through practical work and shared knowledge.
• Explain the business problem behind each feature, not only the technical solution.

Collaboration and Culture