Software Engineer, Applied AI
$175k - $290k • New York • FullTime
Posted 6mo ago
About the job
Auctor is building the AI layer for professional services and software implementation, aiming to automate documentation, discovery, and decision-making for a significant portion of the services work market. As a Software Engineer, Applied AI, you will be instrumental in designing, building, and enhancing the core systems that power our production agents. This role bridges engineering and empirical research, involving work across retrieval, document understanding, tool use, context management, prompting, and orchestration. You will be responsible for shipping new capabilities, analyzing production traces, designing evaluations, and identifying system failures, with a focus on building robust agent systems and making informed product and architecture decisions based on real-world behavior.
Responsibilities
- Build and improve core agent systems, including retrieval, tool use, document understanding, memory, and orchestration.
- Design evaluations and experiments to assess agent quality in production.
- Translate production traces, failures, and user behavior into product and architecture decisions.
- Collaborate with operations, GTM, and deployed teams to understand real workflows and agent breakdowns.
- Evaluate models, prompts, and system designs for real enterprise tasks.
- Manage the entire development cycle from idea generation to implementation, measurement, and iteration.
Requirements
- Strong engineering fundamentals with the ability to ship production systems.
- Fluency in Python.
- Experience building or working with LLM-powered products, agent systems, or related applied AI systems.
- An empirical mindset, utilizing logs, traces, experiments, and usage data.
- Strong systems understanding, recognizing the interplay of retrieval, prompting, memory, tools, and UX.
- High sense of ownership and comfort working in ambiguous environments.
- Well-defined opinions on effective agent system design.