Auctor
Software Engineer, Applied AI
About this role
Join Auctor to build production AI agent systems that power professional services workflows. You'll own the full cycle from design through iteration, working across retrieval, tool use, and orchestration while using production data to drive architectural decisions.
What you'll do
- Build and optimize core agent systems including retrieval, tool use, document understanding, and orchestration
- Design experiments and evals to measure agent quality in production environments
- Analyze production traces and user behavior to inform product and architecture decisions
- Partner with operations and field teams to understand real workflows and failure modes
- Evaluate models, prompts, and system designs on enterprise tasks
- Own projects from conception through implementation, measurement, and iteration
What they're looking for
- Python programming
- LLM-powered product development
- Agent systems architecture
- Production systems engineering
- Empirical experimentation and analysis
- Retrieval and search systems (preferred)
- Evaluation design for AI systems (preferred)
- Workflow automation and tool design (preferred)
Benefits
- Competitive salary ($175,000–$290,000 base)
- Early-stage equity
- Catered lunch and dinners
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Auctor
Auctor builds AI-powered systems that automate professional services workflows, combining agentic AI, LLMs, and full-stack engineering to process unstructured data and power enterprise operations. The company is hiring Software Engineers, Interns, and Forward Deployed Engineers to develop production AI agent systems, design enterprise interfaces, and implement custom integrations for strategic customers.
View all jobs at AuctorLikely interview questions
- Walk us through a time you built or worked on an LLM-powered product or agent system. What went wrong, and how did you diagnose the problem?
- Describe your approach to designing evals or experiments for an AI system. How do you decide what to measure and when you have enough signal to make a decision?