AI Engineer – Agentic Systems

Valority is building Luke, an autonomous AI decision engine for marketing and growth teams. Luke connects fragmented marketing, funnel, CRM, and revenue data, understands business context through a semantic operating layer, and delivers trusted, explainable, financially grounded insights and recommendations. We are looking for an AI Engineer with hands-on experience building production-grade agentic systems. This role is ideal for an engineer who can move beyond prototypes and build reliable AI workflows that reason over enterprise data, execute analytical tasks, and produce explainable outputs users can trust.

What You'll Do

• Build and maintain production-grade AI agents and agentic workflows for Luke's decision engine.
• Design multi-step agent reasoning flows, including planning, tool calling, state management, retries, error handling, and workflow orchestration.
• Implement RAG and semantic retrieval capabilities across structured and unstructured data sources.
• Build and optimize LLM pipelines for analytical reasoning, root-cause analysis, and financial impact estimation.
• Develop evaluation frameworks to measure agent reliability, accuracy, and output quality.
• Collaborate with product and domain experts to translate business logic into agent reasoning and decision workflows.
• Contribute to the architecture of Luke's agentic core, including memory, context management, and tool integration layers.
  
  

• 5+ years of software engineering experience, with at least 1–2 years focused on LLM-based or agentic systems in production.
• Strong Python skills and experience building production AI/ML pipelines.
• Hands-on experience with agentic frameworks (e.g., LangGraph, LangChain, AutoGen, CrewAI, custom agent loops).
• Experience with RAG architectures, vector databases (e.g., Weaviate, Pinecone, pgvector), and semantic search.
• Solid understanding of LLM behavior, prompt engineering, and techniques for improving reliability (e.g., structured outputs, chain-of-thought, tool use).
• Experience integrating with APIs and building tool-use systems for agents.
• Strong debugging and testing skills, especially for non-deterministic AI systems.
  
  

• Experience in analytics, BI, or marketing/revenue data domains.
• Familiarity with SQL generation, data reasoning agents, or structured data pipelines.
• Experience with LLM evaluation frameworks (e.g., RAGAS, TruLens, custom evals).
• Background in MLOps, observability, or production AI system monitoring.
  
  

• Work on a technically ambitious problem: autonomous decision intelligence for enterprise teams.
• Build systems that actually move the needle for real businesses.
• Join a small, high-conviction team where your work shapes the product and architecture.
• Be part of the foundational build of Luke's agentic core.