Plan retry-after and backoff

Prompt that plans Retry-After, exponential backoff with jitter and a hard retry budget. Paste it into an AI tool of your choice, which then plans retry-after and backoff.

Topic: Apps & prototypes
Type: Prompt
Level: Intermediate
Time: about 15 min

Results depend on the tool, the model and what you enter.

Result preview

Example: written by our editors, not real AI output.

API checkout, errors 429/503 Today: retry 5× fixed 200 ms Retry-After: honour header Else: exp backoff + jitter, budget 3 Stop on 400/401/403 Dash: retry_rate, amplified_rps Abort if retries > 15 % traffic

Fill in

What you type stays in this browser window. We store and send nothing.

ServiceWhat fails?

FailureTypical errors?

ClientWho retries?

Your prompt

Help me plan retry-after and backoff for failed calls.
Service: [Service]
Failure: [Failure]
Client: [Client]
Steps: 1. Read which errors are safe to retry and which are permanent. 2. Define Retry-After handling when the server sends it. 3. Propose exponential backoff with jitter and a hard retry budget. 4. Separate client retries from server load shedding and from idempotent writes. 5. Name dashboards for retry rate and amplified traffic. 6. Write a short rollout and abort rule.
Output: retry classes, backoff rule, budget, dashboards and rollout.

Take particular care to retry only what is safe, honour Retry-After and cap the budget.

Before you answer, check: Are all instructions met? If key information is missing or a placeholder in square brackets is still in the text, ask instead of guessing. Mark assumptions as assumptions.

This is an instruction for an AI, not the result itself.

Where this prompt works

These AI tools are a good fit for this prompt:

How to use it

Fill in the fields or tap a suggestion. Leave out names, addresses and other personal data.
Copy the finished prompt into an AI tool of your choice.
Paste error codes, current retry code and traffic peaks, without secrets.

Why does this work?

Blind retries on 503 turn a short overload into a longer outage. Retry-After plus jitter spreads the retry wave.

Tips

Cap total retries per request and stop on 4xx that are not 429.

You may use our prompts freely, including at work. Please do not republish them as your own collection. We give no warranty for AI results; check them yourself. Terms of use

Plan retry-after and backoff

Result preview

Fill in

Your prompt

Where this prompt works

How to use it

Why does this work?

Tips

Also a good fit

Choose pagination for an API list New

Write an API reference people can use right away New

Plan a throttled data backfill New