Coursera and Udemy are now one company, helping more than 300 million learners and 12,000+ enterprise customers build skills for a world being reshaped by AI.
Responsibilities
Build and Design AI systems by selecting the right combination of data, including retrieval pipelines, agentic workflows with tool use, and structured extraction from unstructured sources.
Architect and implement Retrieval Augmented Generation (RAG) pipelines with vector databases (Pinecone, Weaviate).
Build and maintain agentic AI workflows using frameworks such as Langgraph, Mastra, LlamaIndex, CrewAI, or AutoGen.
Design and run evaluation for every AI system, covering accuracy, hallucination and grounding checks, regression suites, and human review loops.
Work with large-scale structured and unstructured datasets on cloud-native databases & storage (S3, Postgres, GCS, BigQuery, Databricks).
Build rapid AI prototypes and experiment with different models, retrieval strategies, and prompts.
Partner with product managers, data engineers and backend/frontend engineers to translate business problems into well-scoped AI solutions.
4+ years of experience in Data Science, Applied AI, or Machine Learning.
Experience taking an AI use case from data exploration and experimentation through a validated working prototype.
Hands-on experience with machine learning and/or modern Generative AI techniques such as LLMs, embeddings, RAG, semantic search, NLP, recommendation, or agentic systems.
At least 2 production deployments involving LLM-based systems.
Experience with AI experimentation and evaluation.
Experience using Python and SQL to work with data.
Experience with Generative AI platforms such as Vertex AI, Bedrock, Azure AI, OpenAI/Anthropic APIs, Hugging Face, LangGraph, LangChain is preferred.
Experience with AI evaluation techniques such as golden datasets, LLM-as-judge, regression evaluation is preferred.
Experience working with modern data platforms such as Databricks, BigQuery, Snowflake, Spark is preferred.
Familiarity with AI observability, model monitoring, responsible AI, PII handling, prompt injection mitigation, and AI guardrails is preferred.