LLM Engineer
On-siteNew York City, New York, United States or New York, United States
Job Summary
Build and iterate on LLM-powered agents that decide production, sourcing, and logistics for global CPG brands. Design robust prompts, tool definitions, and multi-step agent flows to handle messy supply chain data. Select and evaluate LLMs based on latency, cost, and accuracy across the workflow. Ship features from concept to production, owning the full pipeline from API integration to tested systems. Maintain evaluation frameworks and set up observability to track production accuracy and continuously improve performance. Requires 3+ years shipping LLM systems in production, proficiency with modern LLM APIs, and strong system-design skills for agent architecture. Visa sponsorship available; on-site in New York, NY.
Required Qualifications
- 3+ years of hands-on experience shipping LLM-powered features or systems in production environments (not just demos or prototypes)
- Proficiency with modern LLM APIs such as OpenAI, Anthropic, DeepSeek, Gemini, or similar, with a strong intuition for model selection tradeoffs
- Experience building LLM-powered agents or automations that are core to production systems
- Strong system-design skills for tools/functions, structured outputs, and multi-step agent flow architecture
- Experience designing and maintaining evaluation pipelines for prompts, workflows, and LLM outputs
- Hands-on experience with LLM observability, monitoring, and logging tools to track latency, accuracy, and quality in production
- Practical experience with the HuggingFace ecosystem in real projects
- On-site in New York, NY, United States
Desired Qualifications
- Bonus: experience running self-hosted LLMs in production or serious prototypes
- Bonus: background in supply chain, logistics, or operations domain systems
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.