Note: The job is a remote job and is open to candidates in USA. Wand AI is a company focused on integrating AI into the workforce, enabling a hybrid collaboration between humans and AI agents. They are seeking a Staff Software Engineer to build AI systems for governance that analyze organizational data and flag policy violations, with a focus on autonomy and ownership in project execution.
Responsibilities
- Design and build AI and agentic systems that analyze organizational data to catch policy violations, compliance risks, and governance issues, often from ambiguous, non-deterministic signals
- Build agents and pipelines that use LLMs to reason over large volumes of data, going well beyond deterministic, rule based checks
- Architect and build knowledge graph systems that model organizational structure and relationships, and make that context usable by agents
- Take a fuzzy, undefined problem ("find policy violations across messy data"), propose a real technical approach, defend it, then build it with minimal oversight
- Bring engineering rigor to inherently fuzzy territory: testing, evaluating, and iterating on how well your agents actually perform
- Partner with the rest of the Org Intelligence team to get AI-driven insight into the product's governance and compliance surfaces
- Contribute across the stack when it helps, though the core of this role is the AI and agent layer, not the UI
Skills
- A track record of building AI systems as a creator, not a consumer. You can talk in real depth about an agentic workflow, model, or system you built and shipped, not just a tool you use day to day
- Experience building or working with knowledge graphs in a real, shipped system
- Experience building agents or LLM based systems that reason over unstructured or ambiguous data, not simple deterministic pipelines
- Comfortable owning a project from architecture to delivery with real autonomy: you propose the design, defend it, then build it
- Strong software engineering fundamentals and the independence to thrive in a fast moving, remote first, senior-heavy team
- A history of turning fuzzy, non-trivial problems into concrete, working systems
- Practical experience with knowledge graph tooling: graph databases and query languages such as Neo4j/Cypher, RDF/SPARQL, Amazon Neptune, or similar
- Hands on experience with LLM provider APIs (OpenAI, Anthropic, or similar) and agent frameworks such as LangChain or LangGraph
- Comfortable with retrieval infrastructure like vector databases (Pinecone, Weaviate, pgvector, or similar) for grounding agent reasoning in organizational data
- Strong in a language well suited to graph construction, agent pipelines, and data analysis
- Experience building security, risk, or compliance related tooling, especially anything involving policy or violation detection
- Prior experience at a company building agentic or AI-native products, rather than bolting AI onto an existing product as a feature
- Comfortable working fully remote, across time zones, on a small and highly autonomous team
Company Overview