Prompting Reasoning Models Well
Know when a reasoning model is worth the wait and how to brief it for hard, multi step problems.
A taste of a lesson
I gave a reasoning model a staff rota problem and wrote out all the steps it should follow. The answer was worse than expected. Why?
Prescribing every step often backfires with reasoning models. They tend to do better when you describe the problem completely and define success, then let them find the approach. Rewrite the prompt with three parts: all the facts and constraints, what a correct rota must satisfy, and the output format. Add: 'After the table, check each constraint and mark it met or not met.' Then verify two constraints yourself. Which constraint matters most to you, so we check that one first?
Written by the teacher as an example. In your lesson the tutor answers your own questions, and like any AI it can be wrong.
What you will be able to do
- Decide when a reasoning model is worth the extra time and cost
- Brief a reasoning model with goals, constraints and success criteria
- Request stated assumptions and constraint checks in the answer
- Verify reasoning model output independently before relying on it
Lesson plan
- 1 What reasoning models do differently Understand extended reasoning and how products present it differently. Start
- 2 Choosing the right model for the task Match model type to problem type and stakes. Start
- 3 Briefing instead of micromanaging Write prompts that give the full problem and the definition of success. Start
- 4 Built in self checks Make the answer show whether it meets every constraint. Start
- 5 Verifying the result Confirm answers independently, especially when stakes are high. Start
- 6 Building a reasoning playbook Record which problems benefit and which prompt patterns work best. Start
Try asking
About this tutor
Many assistants now offer reasoning models or a reasoning mode that spends extra time working through a problem before answering. They can do noticeably better on multi step analysis, planning, maths and tricky logic, but they are slower, often cost more, and need a different style of prompt. This tutor teaches when to use them, how to brief them with goals, constraints and success criteria rather than step by step instructions, and how to check their answers, since a long chain of reasoning can still reach a wrong conclusion. For experienced users who already prompt well and want to handle harder problems.
Reviews
Students can review a tutor after a paid lesson. Nobody has yet.
About the teacher
Prompt workflows for heavy users: chaining, standing instructions, long documents and reasoning models
9 tutors 269 lessons taught Sample
I work with people who already use AI assistants every day and want more dependable results. My background is in operations and process design, which taught me to treat a prompt like a small procedure: inputs, steps, checks and a clear output. I teach chaining, reusable instructions, long document work and how to test whether a prompt change actually helped....
See Darius's profile and tutorsMore like this
Other tutors on the same or nearby topics.