Backend Engineer, Klue

• 3+ years of backend engineering experience, ideally in search, retrieval systems, or high-scale APIs
• Strong programming skills in Python, Ruby, or similar backend languages
• Experience with search infrastructure (Elasticsearch, OpenSearch, Vespa) and vector search systems
• Understanding of retrieval pipelines, dense retrieval, and hybrid search
• Familiarity with real-time data pipelines (Kafka, Pub/Sub) for indexing workflows
• Experience with distributed systems and microservices, with a focus on reliability and performance
• Familiarity with cloud infrastructure (AWS, GCP, Azure) and container orchestration (Kubernetes)
• Ability to work collaboratively with ML Engineers, understanding their experimentation workflows and constraints
• Strong debugging and profiling skills for production systems

Desirable

• We're looking for a Backend Engineer to join our team in Toronto to work on LLM-powered search and retrieval agents
• In this role, you will build and optimize the retrieval infrastructure, APIs, and data pipelines that power our ML team’s agentic workflows, enabling fast, accurate, and scalable retrieval for advanced user-facing systems
• You will work closely with machine learning engineers, ensuring our backend systems can support rapid experimentation while remaining production-ready
• Design, implement, and maintain retrieval infrastructure and APIs that interface seamlessly with LLM-based agent workflows
• Integrate dense retrieval, hybrid retrieval, and re-ranking models into live systems
• Optimize latency, scalability, and throughput of retrieval systems for real-time agentic pipelines
• Build and maintain vector database infrastructure (FAISS, Milvus, Weaviate, Pinecone, PGVector) and traditional search engines (Elasticsearch, OpenSearch)
• Support retrieval-augmented generation (RAG) workflows, including efficient query execution, chunk retrieval, and caching strategies
• Develop monitoring and observability tools for retrieval pipelines to ensure reliability and transparency
• Work with ML engineers on data pipelines for indexing and re-indexing, enabling continuous improvement of search relevance
• Contribute to the architecture of multi-step retrieval agents, ensuring clean abstractions between the backend and ML layers
• Take ownership and run with ambiguous problems
• Jump into new areas and rapidly learn what's needed to deliver solutions
• Bring scientific rigor while maintaining a pragmatic delivery focus
• See unclear requirements as an opportunity to shape the solution
• LLM platforms: OpenAI, Anthropic, open-source models
• ML frameworks: PyTorch, Transformers, spaCy
• Search/Vector DBs: Elasticsearch, Pinecone, PostgreSQL
• MLOps tools: Weights & Biases, MLflow, Langfuse
• Infrastructure: Docker, Kubernetes, GCP
• Development: Python, Git, CI/CD