Applied AI Engineer

Responsibilities:

• Take models and methods from research and make them production services.
• Build the evaluation that decides whether a change is actually an improvement.
• Tune retrieval, prompting and memory for accuracy and for cost.
• Instrument what ships, so a regression is visible before a customer finds it.
• Work across the line in both directions — back to research with what broke in production.

Requirements:

• A working understanding of modern ML and LLMs — what they do, where they fail, and why.
• Strong software engineering: you write code others can run and change.
• 2+ years shipping something real, ML-backed or not.
• Python, and comfort with PyTorch or an equivalent framework.
• The judgment to tell a genuine improvement from a better-looking number.

Nice to have:

• Retrieval, embeddings or vector search in production.
• Evaluation frameworks and LLM-as-judge setups, and their limits.
• Fine-tuning, LoRA or adapters.
• Having owned a service as well as a model.

Skills

LLM, LLMs, PyTorch, Python