Description
You will design and deliver AI applications, focusing on LLM integration, retrieval systems, and evaluation pipelines. You own the full lifecycle from system design to MLOps implementation.
This role is on-site.
Responsibilities
- Build and integrate applications with LLMs using vector, graph, or structured retrieval approaches.
- Design practical systems with clear tradeoffs, implementing MLOps patterns and context management.
- Create evaluation pipelines combining quantitative and qualitative methods to align metrics with user workflows.
- Translate technical findings into clear guidance for other engineers regarding AI engineering practices.
- Fine-tune and evaluate models using PyTorch and manage lightweight training processes.
Required Skills
- 5+ years of hands-on programming experience in Python.
- Proficiency in at least one additional language: Swift, Ruby, or JavaScript.
- Experience with LangChain, MCP, or similar model integration frameworks.
- Strong understanding of MLOps, retrieval strategies, and context management.
- Ability to design and execute experiments with clear hypotheses and success criteria.
- Bachelor's degree in a relevant field or equivalent applied data science experience.
Preferred Skills
- Familiarity with the Hugging Face ecosystem (Transformers, Datasets, evaluation libraries).
- Experience rapidly applying new model capabilities and emerging AI research.