AI Query Intent Classification Engineer

Quelion Cloud

Remote · $100k – $130k / year · Full-time · Posted 4 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

The same word means different things in different contexts. 'Python' in a developer tool search means the programming language. 'Python' in a nature content platform means the snake. A search engine that doesn't understand query intent before retrieving results will surface confidently wrong answers to both users. Our search platform powers the discovery experience for 90 enterprise SaaS products. Query intent classification — understanding what type of result the user wants, what domain they're searching in, and what action they're trying to take — sits upstream of retrieval and fundamentally affects result relevance. We have a basic intent classifier; it's not good enough for the diversity of query types our clients see. You'll redesign the query intent classification system: building a multi-label classifier that identifies query type, domain, and action intent simultaneously, implementing few-shot adaptation that lets each client's search engine learn client-specific intent taxonomies with minimal labeled data, and building the evaluation framework that measures intent classification accuracy against relevance outcomes rather than just label agreement.

Fully remote.

Responsibilities

Redesign multi-label query intent classifier handling type, domain, and action intent dimensions
Implement few-shot intent adaptation for client-specific query taxonomy customization
Build evaluation framework connecting intent classification accuracy to search relevance outcomes
Develop intent-aware retrieval routing directing queries to appropriate index shards
Create intent distribution monitoring detecting when new query types emerge in production
Fine-tune base intent models on client query logs with privacy-preserving aggregation

Requirements

3+ years in NLP engineering with text classification or search relevance experience
Strong Python and PyTorch skills with HuggingFace model fine-tuning experience
Understanding of information retrieval and search quality evaluation
Experience with few-shot or zero-shot learning approaches for low-resource classification
Ability to design evaluation frameworks that measure real user outcome impact