About the job
Klue develops competitive intelligence tools that help organizations turn information into actionable insights. The team is expanding its Engineering group in Toronto, with a focus on practical AI applications that support insight generation.
Role overview
The AI Software Engineer will work on developing and refining large language model (LLM)-powered agents. This position centers on building systems for reasoning, planning, and workflow automation, as Klue advances its approach to generating insights. The work involves hands-on development with modern AI technologies, especially for those with strong backend and machine learning experience.
Key responsibilities
- Design retrieval-augmented generation (RAG) systems to improve query understanding, document retrieval, and response synthesis using agentic workflows.
- Build and optimize retrieval pipelines, including BM25, dense retrieval, hybrid retrieval, and re-ranking methods.
- Create evaluation frameworks for retrieval and generation, focusing on offline metrics such as recall, MRR, nDCG, and human-in-the-loop assessments.
- Experiment with query rewriting, expansion, and classification to enhance retrieval relevance.
- Collaborate with Product teams to launch machine learning-powered search agents in production settings.
- Monitor, debug, and optimize latency, accuracy, and scalability for retrieval and generation components.
- Contribute to the design and implementation of data pipelines for training retrieval and ranking models, including dataset curation, augmentation, and labeling workflows.
- Stay current with new developments in LLMs, retrieval strategies, and agent architectures, and assess their relevance for Klue’s systems.
Requirements
- Experience in software engineering.
- Familiarity with information retrieval systems, search relevance, and ranking models.
- Strong background in machine learning and backend technologies.
- Ability to work well within a team.

