reliable · advanced · 45 min

Many workers, one dispatcher

Process a batch of claims in parallel with a cap on how many run at once, retries for the ones that fail, and no claim lost.

Start step 1 ↓

HANDS-ON LAB · YOU RUN IT ON YOUR MACHINE · 4 STEPS

Step 1 of 4

Make 20 claim emails and process them one after another, timed

Run this to check

time uv run --with openai --with pydantic fanout.py claims/ --max 1

You pass when

20 results and a total time. This is the number to beat.

Files you write
- claims/
- fanout.py
Step 2 of 4

Run them four at a time

Run this to check

time uv run --with openai --with pydantic fanout.py claims/ --max 4

You pass when

The same 20 results in well under the first time. The log never shows more than four in flight.

Files you write
- fanout.py
Step 3 of 4

Make one call in five fail at random, and retry twice with a growing wait

A batch that reports 18 of 20 with no word on the other two is the failure that reaches customers.

Run this to check

uv run --with openai --with pydantic fanout.py claims/ --max 4 --chaos 0.2

You pass when

Every claim ends as done or as failed after three tries. The count of both adds up to 20.

Files you write
- fanout.py
Step 4 of 4

Print a summary: done, retried, failed, total cost

Run this to check

uv run --with openai --with pydantic fanout.py claims/ --max 4 --chaos 0.2 --summary

You pass when

Four numbers, and the ids of any claim that failed for good.

Files you write
- fanout.py

No transcript here on purpose. This page does not show terminal output, because model answers differ from run to run. Each step says what a pass looks like; your own terminal is the proof.