Field Guide · For engineering students

Your last 7 days: a revision plan from your topic list.

Field Guide, Exam prep with AI, part 1 of 6. A 7-day revision plan from your own topic list, tested on free ChatGPT: what worked and the catch.

1 Nov 2026 · 3 min read · sources read up to 1 Nov 2026

Seen on Instagram: the post this article goes with.

Seven days to go, nine topics on the list, and four of them still make your stomach drop. A plan helps, but making one takes the energy you would rather spend revising. This is part 1 of the Field Guide series on exam prep with AI: one job, tested on a real screen, with the catch said plainly.

The job, and what to type

Write down the topics you still need to revise, in your own words, and mark each one from 1 (weak) to 3 (strong). Then ask for a plan with your limits in it. The topic list below is a sample, made up for this demo. Use a list you typed yourself, never a question paper or anyone else's material.

My exam is in 7 days. Below are the topics I still need to revise, with how sure I feel about each (1 = weak, 3 = strong).

Make me a 7-day revision plan:
- Put the weak topics earlier and give them more time.
- I can study 4 hours a day, no more.
- Keep day 7 light, for a quick recap of everything.
- Show it as a table: day, topics, hours, what to do.
- If something doesn't fit, tell me instead of squeezing it in.

1. Arrays and strings - 3
2. Linked lists - 2
3. Stacks and queues - 2
4. Trees - 1
5. Heaps - 1
6. Hashing - 2
7. Graphs (BFS and DFS) - 1
8. Sorting - 3
9. Dynamic programming - 1

Why these lines

"Make me a revision plan" gets you a plan. What makes it your plan is the rating: marking each topic 1 to 3 for how sure you feel tells the AI which topics scare you, so it can put them first and give them more time. The hour limit keeps the week real, and "if something doesn't fit, tell me instead of squeezing it in" asks it to say so rather than quietly cramming.

How it was tested

One run on 1 November 2026, on free ChatGPT in a web browser, signed in to a personal demo account. The chat was a temporary chat set to Unpersonalized, which does not use memory, custom instructions or plugins ^[2]. The five checks were written down before the prompt was sent.

1. Every one of the 9 topics is in the plan, and nothing is added.
2. The 4 weak topics come earlier and get more time than the strong ones.
3. No day goes over 4 hours.
4. Day 7 is light, for a recap.
5. It says plainly if something doesn't fit.

What came back

A full table arrived in about ten seconds. Trees and Heaps on day 1, Graphs on day 2, Dynamic programming on day 3, then the medium and strong topics, mixed practice on day 6 and a 2-hour recap on day 7. Every day stayed at 4 hours or less, 26 hours in all, and it kept the last 2 hours free as a buffer instead of filling them.

Checks 1, 3, 4 and 5 passed.

The catch

Check 2 failed, in two ways. The answer opened with "You have 9 topics to revise, with 5 weak topics rated 1/3". The list has 4. And inside day 1, Heaps, a weak topic, got 1.5 hours, while Arrays and strings and Sorting, both strong, got 2 hours each. The plan looks tidy, so neither slip jumps out.

The fix takes two minutes. Count the topics in the plan against your own list, then add up the hours per topic, not just per day. Move time to any weak topic that ended up short.

The stamp

Half works · 1 Nov 2026 · free ChatGPT, in a browser · 4 of 5 checks passed. Why half: the week itself was right (all nine topics, no day over 4 hours, a light last day, a plain note about the spare time), but it miscounted the weak topics and gave one of them less time than two strong ones. A usable week, quickly, from your own list. Check the count and the hours before you follow it. One run is one run: try it with your own list and see what yours gets wrong.

Before you use it

For revising only. Never in an exam, and for marked work, follow your college's AI rules. ChatGPT is for ages 13 and over, or the minimum age in your country, and under 18 needs a parent's or guardian's permission ^[1].

The raw answer, word for word

You have 7 days × 4 hours = 28 total study hours. You have 9 topics to revise, with 5 weak topics rated 1/3, so the plan prioritizes those early while leaving your stronger topics for shorter practice sessions.

I'll keep Day 7 light and reserve it for a quick recap of everything.

Your 7-day revision plan

| Day | Topics | Hours | What to do |
| Day 1 | Trees (1), Heaps (1) | 4 h | Trees — 2.5 h: traversals, BST operations, height, common problems. Heaps — 1.5 h: min/max heaps, insert/delete, heapify. Solve practice questions for both. |
| Day 2 | Graphs: BFS & DFS (1) | 4 h | Graphs — 4 h: adjacency lists/matrices, BFS, DFS, visited arrays, connected components, and traversal problems. |
| Day 3 | Dynamic programming (1) | 4 h | DP — 4 h: identify states and transitions, base cases, memoization vs tabulation, and classic problems such as climbing stairs, knapsack, and longest common subsequence. |
| Day 4 | Linked lists (2), Stacks & queues (2) | 4 h | Linked lists — 2 h: reversal, cycle detection, insertion/deletion. Stacks & queues — 2 h: implementations, monotonic stacks, balanced brackets, and queue operations. |
| Day 5 | Hashing (2), Arrays & strings (3) | 4 h | Hashing — 2 h: maps, sets, collisions, frequency counting. Arrays & strings — 2 h: two pointers, sliding window, prefix sums, and common string problems. |
| Day 6 | Sorting (3), mixed weak-topic practice | 4 h | Sorting — 2 h: merge sort, quicksort, complexity, stability. Mixed practice — 2 h: revisit your weakest problems from trees, heaps, graphs, and DP. |
| Day 7 | Quick recap of all 9 topics | 2 h | Review summary notes, complexities, templates, and mistakes. Spend roughly 10–15 minutes per topic, then finish with a short confidence check. Keep the rest of the day free. |
|  | Total scheduled | 26 h | 2 hours remain unallocated rather than forcing extra study into the plan. |

What to prioritize each day

Days 1–3: Build your weakest foundations

Focus on understanding and implementing the core ideas. Don't spend all your time watching tutorials; write code and solve problems from memory.

Days 4–5: Practise patterns

Aim for a mix of easy and medium questions. Track the mistakes you make and revisit them the following day.

Day 6: Test exam readiness

Attempt timed questions without notes. Use the final two hours to repair gaps in your weakest topics rather than studying everything equally.

Day 7: Keep it light

Review formulas, time complexities, algorithm templates, and your error log. Avoid starting unfamiliar material or doing a long mock exam.

What doesn't fit?

All nine topics fit into the plan, but 26 hours are explicitly scheduled, leaving 2 hours unused. That's intentional: you can use them as a buffer if a weak topic takes longer than expected, or simply preserve the time for rest.

One important adjustment: if dynamic programming or graphs feels especially difficult after Day 3, use some of the Day 6 mixed-practice time to revisit it.

Your rule for the next 7 days: understand the basics, solve problems without help, and review mistakes. Aim for mastery of common exam patterns rather than trying to learn every possible variation.

ChatGPT can make mistakes. Check important info.

Next in the series, part 2: practice questions from your own notes.