Software Engineer

Position Overview

In this role, you will build the core engine underneath Naboo: high-throughput indexing pipelines, graph traversal algorithms, hybrid retrieval (lexical + vector + graph), and low-latency query execution. This is deep systems and data engineering, not prompt engineering - the hard problems are retrieval precision, latency, and context density at enterprise scale.

Main Responsibilities

• Core Engine & Real-Time Data Pipelines: Design, build, and scale data pipelines, live ETL processes, and indexing engines that continuously pull and unify data from multiple customer integrations into a single knowledge graph, in real time - plus the graph-traversal systems that transform enterprise artifacts into that unified graph.
• Retrieval & Relevance: Optimize hybrid retrieval (BM25, dense embeddings, graph expansion, reranking) to maximize precision and minimize token footprint for downstream agent execution.
• Evaluation & Benchmarking: Build offline and online eval frameworks to benchmark retrieval accuracy, latency, and agent task success against real enterprise workloads.
• Production Ownership: Share end-to-end ownership of core systems - from schema design and implementation to deployment, observability, and production reliability.
• Technical Direction: Collaborate closely with the team on architectural decisions as agent paradigms (MCP, multi-agent coordination, context caching) rapidly evolve.

Requirements

• 3-5 years of experience building high-performance backend systems, distributed data pipelines, or data-intensive applications.
• Practical experience with retrieval & search: embeddings, vector search, lexical search (BM25 / Lucene / OpenSearch), and reranking trade-offs.
• Agentic SDLC practitioner: hands-on experience working in an AI-native development environment - actively using AI coding agents (Claude Code, Cursor, etc.) in your daily engineering workflow, designing with agent-first primitives, and dogfooding the paradigm we build for our customers.
• Product velocity: thrives in a fast-paced environment with high ownership, taking ambiguous technical problems from 0 to production. We're looking for people who identify problems and propose solutions on their own, rather than waiting to be told what to do. We encourage people to bring different ideas and value people who are not afraid to voice their opinion.

Nice to Have

• Experience with graph databases (Neo4j, Memgraph, FalkorDB) or knowledge graph representations.
• Familiarity with the Model Context Protocol (MCP) or agentic execution frameworks.
• Background in search relevance, learning-to-rank, or synthetic evaluation generation.
• Early-stage startup experience.

• B.Sc. in Computer Science, or equivalent hands-on experience at strong engineering companies.