Software Engineer, AI

Klue•Toronto, ON

5d•Hybrid

About The Position

Klue Engineering is hiring a Software Engineer with production experience to join their Toronto AI engineering team. The role involves building and improving LLM-powered agent systems that transform competitive data into actionable intelligence for customers. This is a hands-on role where the engineer will ship code into the agent stack, pair with senior engineers, and gradually take on more ownership. The ideal candidate has shipped a backend system in a prior role and is eager to specialize in retrieval, RAG, and agent infrastructure.

Requirements

Production software engineering experience, including shipping customer-facing features with code review, deploys, and on-call responsibilities.
Strong Python fundamentals, including modern practices like type hints, async/await, FastAPI or similar, SQLAlchemy or another ORM, and pytest.
Curiosity about LLMs, RAG, and agentic systems, with experience in side projects, courses, or prior work with LLMs.
Solid CS fundamentals, including understanding of HTTP, databases, basic distributed systems concepts, and tradeoffs in building reliable services.
Comfort with cloud and containerized environments, including Docker and deployment to AWS/GCP/Azure.
Experience using AI coding tools like Copilot, Cursor, or Claude Code.
Customer-oriented mindset, focusing on the purpose and users of the developed features.
Ownership and growth mindset, taking projects to completion, seeking and acting on feedback, and proactively finding new responsibilities.

Nice To Haves

Hands-on experience with Elasticsearch / OpenSearch, vector DBs (Pinecone, PGVector, FAISS, Weaviate), Temporal, Pydantic-AI, fastmcp, or LangChain.
Exposure to MLOps tooling (Logfire, Langfuse).
A side project or open-source contribution involving LLMs, retrieval, or agents.
Experience writing evaluations for LLM systems.
Background in information retrieval, search, or NLP from school or work.

Responsibilities

Ship backend code that powers our agents.
Build and modify agent services, workflows, and MCP tools.
Help improve retrieval and ranking by pairing with senior engineers on hybrid retrieval, query rewriting, chunking strategies, and re-ranking.
Write and run evaluations to assess agent performance.
Improve LLM-powered workflows by tweaking prompts, adjusting retrieval parameters, adding caching, reducing latency, and measuring production impact.
Ensure reliability by writing tests, monitoring deploys, following up on errors, and maintaining code quality.
Collaborate with senior ICs, the product team, and other engineers across infra and data, asking questions, getting feedback, and sharing knowledge.