Red Belt · Promotion exam · LAD
LLM API Developer
Validates that you can implement application features with LLM APIs and manage quality, cost and latency using structured output, tool calling, streaming and caching.
Daily practice · Free
10 questions · 20 min · No sign-up needed
Promotion exam · Proctored
30 questions · 50 min · Raw score of 70 or more · $3.99 / one attempt
Promotion exam detailsExam overview
- Questions
- 30
- Time
- 50 min
- Pass mark
- 70 pts
- Recommended experience
- At least 1 year of experience developing HTTP APIs
Every question has five choices and is marked on an OMR answer sheet. Multiple-choice and true-or-false questions ask for the right combination from a box; ordering and matching questions ask for the right arrangement. Judgment questions (L3) are worth 3 points and all others 2; the raw score is converted to a 100-point scale and graded 1 to 9.
Exam domains
1. API basics and message design20%
2. Structured output and tool calling25%
3. Performance and cost optimization20%
4. Reliability and error handling20%
5. Evaluation and operations15%
Weights follow the Promotion exam. Practice and mock exams may mix domains differently.
What this exam verifies
- Get the intended response reliably through message structure and parameters
- Connect models to the application with structured output and tool calling
- Reduce latency and cost with streaming, batching and caching
- Handle errors, rate limits and retries robustly
- Maintain quality through evaluation and monitoring
Who it is for
- Backend and full-stack developers adding LLM features to products
- Engineers moving AI feature PoCs into production
Does not cover model pre-training or operating GPU infrastructure.
Domain details
Domain 1. API basics and message design
- Knowledge: Message roles, Tokens and maximum output, Sampling parameters such as temperature, Model selection
- Skills: Choose the model and parameters that fit the requirements
Domain 2. Structured output and tool calling
- Knowledge: Structured output based on JSON Schema, Tool definitions, The tool-calling loop, Parallel tool calls
- Skills: Validate output against the schema, Return tool results to the conversation
Domain 3. Performance and cost optimization
- Knowledge: Streaming, Prompt caching, Batch API, Routing by model tier
- Skills: Arrange prompts to raise the cache hit rate, Weigh latency against cost
Domain 4. Reliability and error handling
- Knowledge: Rate limits, Exponential backoff, Timeouts, Idempotency, Interrupted responses
- Skills: Tell retryable errors apart from the rest, Recover from partial failures
Domain 5. Evaluation and operations
- Knowledge: Evaluation sets, Logging and tracing, Handling personal information, Pinning model versions
- Skills: Run regression evaluations before switching models, Define operational metrics
Official documentation
Anthropic
Red Belt Course · LLM API Developer
Practice, mock exams and every course book are free. Pay only when you take an official test.