LLM-Powered Local Intent Data Scientist

NewsBreak; Mountain View (CA); On-site; USD 190,000 - 260,000; Full time

Benefits offered by this job

Discretionary bonus

Stock options

Job summary

NewsBreak is building the systems that understand what local consumers want, turning raw queries and behavior into structured intent. You will bring LLMs into production to improve local search and recommendations at scale, working with embeddings, retrieval, and NLU to map natural language to local supply.

You will define metrics, run experiments, and ship real features that affect user engagement and discovery across local categories and merchants.

Qualifications

• 5+ years as an Applied Scientist or ML Engineer on query understanding, intent, semantic search, or NLU.
• Production experience shipping models that serve real users.
• Hands-on with embeddings, semantic search, intent classification, taxonomy, or interest-graph construction.
• Experience bringing LLMs into production systems.

Responsibilities

• Build query‑understanding, user‑intent classification, and semantic‑search models for local consumer demand.
• Design and grow intent taxonomies and interest graphs over local categories and merchants.
• Ship LLM‑powered features into production for search and recommendations at consumer scale.
• Work with embeddings, retrieval, and NLU to connect messy consumer language to local supply.
• Define metrics and run experiments that tie intent quality to downstream conversion.

Skills

Query understanding

Intent classification

Semantic search

Production ML

LLMs in production

Embeddings

Python

PyTorch

Tools

Python

PyTorch

Job description

NewsBreak is building the systems that understand what local consumers want, turning raw queries and behavior into structured intent. You will bring LLMs into production to improve local search and recommendations at scale, working with embeddings, retrieval, and NLU to map natural language to local supply.

You will define metrics, run experiments, and ship real features that affect user engagement and discovery across local categories and merchants.