🛠 Tech Stack

💼 About This Role

You'll shape AI infrastructure and drive production-ready LLM experiences at Ruby Labs. You'll own advanced prompt systems, structured outputs, and complex LLM workflows using LangChain or LlamaIndex, making data-driven decisions around model performance and cost.

🎯 What You'll Do

• Design dynamic prompt templates with conditional logic for optimal generation quality.
• Implement structured outputs using JSON mode, function calling, and Zod schemas.
• Build evaluation pipelines with Langfuse to score response quality in real time.
• Perform deep debugging of LLM chains using Langfuse traces to optimize cost and latency.

📋 Requirements

• Deep knowledge of Node.js & Next.js to build reliable services.
• Proven experience in dynamic prompting with input-dependent content.
• Experience with OpenRouter for unified APIs and cost-effective model selection.
• Understanding of LLM observability with Langfuse or similar tools.

✨ Nice to Have

• Practical experience in fine-tuning models for domain tasks.
• Understanding of RAG architecture including indexing and retrieval.
• Basic knowledge of Python for data science scripts.

🎁 Benefits & Perks

• 🏖️ Unlimited PTO
• 💻 Company-provided MacBook
• 🌍 Remote work environment
• 🎉 Paid national holidays

📨 Hiring Process

Estimated timeline: 3-5 days review

1. 1Recruiter Screening· 40 min
2. 2Technical Interview· 60 min
3. 3Final Interview· 30 min