Self-paced
Building with LLM APIs — Start any time
Two days of Python against the OpenAI and Anthropic APIs: streaming, tool use, structured output, retrieval, evaluation and a bill you can predict.
When
Start any time
See the whole scheduleWhere
Start any time
Start the moment you buy it, work at your own pace, and keep access for twelve months.
Asia/Colombo
About this course
Advanced
2 days · 12 hours
Two days of Python against the OpenAI and Anthropic APIs: streaming, tool use, structured output, retrieval, evaluation and a bill you can predict.
What you'll be able to do
- Call the OpenAI and Anthropic APIs from Python and stream a response into a user interface as it arrives.
- Read a response properly: content, stop reason, token usage and error types, and log what billing and debugging need.
- Return output against a JSON schema, validate it, and repair or reject what fails instead of passing it downstream.
- Implement a tool-use loop that defines tools, executes calls, returns results and copes with the wrong tool being chosen.
- Build a retrieval pipeline over a document set with chunking, an embedding model, a vector store, reranking and citations.
- Write an evaluation suite from real failures, combining deterministic assertions with a model-graded rubric, and run it in CI.
What's included
- Class recording, available for 12 months
- Course repository: Python and TypeScript examples, a retrieval pipeline and an evaluation harness
- Course workbook with each API pattern and the failure case that motivates it
- Evaluation set template and a cost and latency tracking sheet
- MyLearnPlus certificate of completion
- Instructor email support after the class for questions on your own implementation
- One free repeat of the same live class, terms shown on the booking page
