Name the task and the strongest relevant alternative.
Project Top Gun
AN ONGOING PROJECT
Choose a contest worth winning, define the measure, and do the work to become the best at it. Ambition needs a test.
There are no points for second place.
PUT IT TO WORK
- Choose one contest and one relevant baseline.
- Keep the inputs, setup, and scoring visible.
- Measure the result and the cost of getting it.
START WITH THE IDEA
The idea in brief.
Be unapologetic about trying to be the best at something specific. Name the task, the strongest relevant alternative, and the evidence that would count as winning. Then let the comparison change what you build.
A claim without a contest
- Say it is better
- Choose flattering examples
- Declare a win
A contest you can measure
- Define task + baseline
- Compare the same work
- Publish result + limits
THE DIRECTION
What we’re working toward.
Measure outcomes, effort, reliability, and cost.
Publish a comparison before claiming a win.
HOW WE THINK ABOUT THE WORK
Principles with practical consequences.
Choose the contest
“Best” needs a task and an audience. Improving an essay draft, checking a calculation, and understanding AI spend are different contests; each needs its own definition of success.
A strong baseline makes the result useful
Compare with something a real user would choose. Keep inputs and conditions comparable, record failures, and state what the comparison leaves out.
Confidence must survive measurement
Ambition gets the work started. Evidence tells you whether it is working. A pleasing demo or a larger model is not a benchmark win; reliability, effort, and cost can change the decision.
A SMALL WORKING PRACTICE
Put the principle to work.
Choose one contest and one relevant baseline.
Keep the inputs, setup, and scoring visible.
Measure the result and the cost of getting it.
FROM IDEAS TO TOOLS
Explore the related work.
Each example gives you a task, something to look for, and a next decision. Product pages describe the current demos, releases, and source availability.
Essay Practice →
Write an argument, get a practice grade, and improve your next draft.
A task to try
Improve an argument through another draft.
What to look for
Practice feedback should help a writer make a specific revision. It is independent practice, not an official test score.
Next decision. Choose a set of essays and an explicit feedback rubric before comparing graders.
Independent GRE®-style practice, not affiliated with ETS. Runs entirely in your browser and works best on a laptop or desktop: the first session downloads about 64 MB. A desktop app is coming later.
Calculator →
Explore basic and scientific calculations, with a graphing workspace.
A task to try
Check a calculation you can independently verify.
What to look for
Accuracy, clear input handling, and understandable results all matter.
Next decision. Keep a small task set with known answers and record effort and failure cases alongside accuracy.
Explore the app with demonstration data.
AI Effectiveness →
See the work behind your AI usage: tokens linked to the conversations, commits and projects they went into.
A task to try
Understand what work AI usage contributed to.
What to look for
Look for the connection between usage, conversations, commits, and projects rather than treating token counts as value.
Next decision. Define a useful outcome first, then compare the usage and effort required to produce it.
Signed and notarized desktop release 0.7.8.
Follow the code.
Published repositories identify their own license and release status. The Open Source and Pro Source pages explain the two paths.
READ IT IN ORDER
A path through the ideas.
The Dogfood Notebook →
From a problem you feel to a tool that earns its place.
Building with AI →
A thesis about specificity, followed by a standard for testing it.
PRINCIPLES & FIELD NOTES
Follow the work.
guide 5 min read
How the GRE® Issue essay is scored (2026)
How the GRE Issue essay is scored: the 0–6 half-point scale, holistic rubric, human raters plus ETS e-rater, and what separates each score band in ETS's guide.
Project Raise the Floor · Project Top Gunguide 5 min readHow to go from a 4 to a 5 on the GRE® Issue essay
A 4 is competent; a 5 is thoughtful. The GRE essay differences in position, reasoning, examples and structure, plus a revision checklist and timed drills.
Project Raise the Floor · Project Top Gunguide 4 min readThe GRE® "two views" Issue prompts, and how to answer them
How to answer GRE Issue prompts that ask which view aligns with your own: what the task requires, a planning template, and four original practice prompts.
Project Raise the Floor · Project Top Gunguide 4 min readThe GRE® Argument essay was removed: what changed
ETS removed the GRE Argument essay on September 22, 2023. Here is the full list of changes in the shorter GRE and what they mean for Analytical Writing prep.
Project Raise the Floor · Project Top Gunguide 4 min readWhat is a good GRE® Analytical Writing score?
What is a good GRE AWA score? ETS percentile ranks for every Analytical Writing level (2022–2025 data), average scores by field, and how programs use the score.
Project Raise the Floor · Project Top Gunguide 4 min readFree GRE® essay practice vs ScoreItNow!
ETS ScoreItNow! and Hopper Labs Essay Practice compared: price, what each includes, how grading works and privacy, so you can pick the right GRE essay practice.
Project Raise the Floor · Project Top Gunprinciple 3 min readFind the right-sized niche
Specialize around a real problem. Generalize when the need and your capabilities justify it.
Project Dogfood · Project Top Gunprinciple 3 min readChoose a contest you can measure
Project Top Gun starts with the discipline to define what winning means.
Project Top Gun · Project Raise the RoofAS THE WORK DEVELOPS
Project updates.
A clearer way to follow the project
The project now has the same compact detail layout and tabs as Research, with principles, practical build examples, reading paths, and connections collected in one place.
The project hub is open
The first reading paths and related tools are collected here. These are the project’s aims; individual product pages describe current availability.
ONE BODY OF WORK
Follow the connections.
Project Dogfood →
Actual preference helps choose a meaningful contest and exposes the moments when another tool wins.
Project Raise the Roof →
Raise the Roof explores what more demanding work becomes possible; Top Gun tests whether a particular approach is better.
Project Raise the Floor →
Making useful tools accessible gives more people a starting point from which to improve and compete.
See the larger picture.
The content graph follows project membership, shared posts, reading paths, topics, and referenced tools. Every connection has a stated reason.
For agents: the public index includes project principles, build examples, dated updates, and related posts.