About the Project

We are building production-grade agentic content workflows for an English Language Learning platform. It uses AI agent architectures to generate and validate educational content at scale, serving multiple products across the organization. The team owns the end-to-end lifecycle: workflow design, LLM orchestration, evaluation, and production operations.

About the Role

We are looking for a senior AI engineer ready to work at the heart of production AI systems. You will contribute directly to two core services: Conversation Brain, an LLM-driven conversation engine, and Ambient ORA, a speech assessment engine powering multiple products for our client. The role sits at the intersection of product engineering and AI science, bridging research outputs from the R&D team into production-grade services.

Key Responsibilities

• Design, develop, and optimize AI-driven content generation workflows.
• Contribute directly to our codebases, building and maintaining agentic workflows using Python (FastAPI, CrewAI, LangGraph, LangChain) and Go where needed.
• Build and maintain agentic workflows using CrewAI, LangGraph, and LangChain.
• Ensure code quality through unit and integration tests written as part of the development workflow, in line with our shift-left approach where developers own test creation.
• Operationalize AI models in collaboration with the AI R&D.
• Implement LLM observability and evaluation using Langfuse, OpenLit, and New Relic.
• Design and run LLM evaluation benchmarks and regression detection.
• Integrate and manage resources across Azure (Container Apps, Service Bus, Blob Storage) and AWS.
• Ensure robust CI/CD pipelines and contribute to SDLC best practices.
• Package, deploy, and scale applications using Docker; Kubernetes experience is a plus.

Required Skills & Qualifications

• Strong Python development skills (FastAPI) and hands-on experience with agentic and LLM frameworks (CrewAI, LangGraph, LangChain, or equivalent).
• Production experience with LLMs, including prompt engineering, model evaluation, and operationalizing research outputs into production services (Azure OpenAI, Google Vertex AI).
• Practical experience with Azure services (Container Apps, Service Bus, Key Vault, Blob Storage) and familiarity with AWS.
• Experience with LLM observability and monitoring tools (Langfuse, OpenLit, New Relic, or equivalent).
• Solid grasp of CI/CD, DevOps, and SDLC. Working knowledge of Docker, with k8s experience as a plus.
• Familiarity with ML/NLP tooling (PyTorch, Hugging Face Transformers, spaCy, scikit-learn) a strong plus.
• Daily user of AI development tools (Cursor, Claude Code, Copilot).
• Fluent English, spoken and written.
• Nice to have: Working knowledge of Go, with the ability to read, modify, and maintain existing Go services.