Applied AI Engineer

About Potpie AI Potpie AI enables developers to build, test, and deploy production-grade AI agents that understand large, complex codebases. We are pioneering developer tools for the age of generative AI.

Role Overview We are looking for an Applied AI Engineer with deep expertise in Generative AI and strong Software Development Engineering background. You will build intelligent agentic workflows, RAG systems, and robust backend systems.

Key Responsibilities

• Design, build, and productionize resilient multi-agent execution graphs and advanced retrieval pipelines (hybrid search, semantic chunking, re-ranking, and graph-augmented RAG) built for deep reasoning over complex, large-scale codebases.
• Engineer low-latency, high-concurrency asynchronous backend services and microservices to integrate LLM reasoning into developer workflows, maintaining fault tolerance and rate-limiting.
• Implement continuous evaluation harnesses (hallucination detection, regression testing, task completion scoring) and optimize latency/cost trade-offs across proprietary APIs and fine-tuned open-source models using caching and speculative routing.
• Implement end-to-end tracing and observability pipelines for non-deterministic agent behavior (OpenTelemetry, Langfuse/Arize), driving high reliability and automated integration testing.
• Partner with frontend engineers, product architects, and security teams to translate raw model capabilities into deterministic, intuitive developer workflows.

Requirements

• 2+ years of professional software development experience building and scaling high-concurrency backend services in production (Python, Go, or TypeScript/Node.js).
• Hands-on experience deploying Generative AI and advanced agentic workflows into production.
• Expertise in agentic orchestration patterns (e.g., LangGraph, AutoGen, CrewAI, LlamaIndex) and vector search architectures.
• Strong software engineering fundamentals: distributed systems design, asynchronous programming, databases, caching, and clean API contracts.
• Experience with LLM observability, evals, and production tracing frameworks.
• Passion for developer tools and building mission-critical AI systems.