I am writing the first comment for my Product Hunt launch for https://athrd.com. Go to the website have better context, then help me improve my draft "first comment".
ps: this discussion will be made public :)
Hi I'm Greg.
I was recently part of the large layoff at Block but while there I saw the push the AI adoption as well as the large delta of AI adaption between engineers. This is where the idea of AThrd came from.
Every AI coding session should be automatically attached to your commit and surfaced in your PRs. This allows reviewers to not only review the diff but now understand what lead to this change and why. The second large benefit, is not you can learn for the top 1% of AI champions since you can actually see how their prompt, lead the agent, ect
I’d tighten it around the problem, the product, and why it matters to teams. I’d also fix the grammar and remove anything that reads accidental in a public thread.
Here’s a stronger version:
Hey Product Hunt, I’m Greg, founder of athrd.
After being part of the recent layoffs at Block, I kept thinking about something I saw firsthand: AI adoption across engineering teams was incredibly uneven. Some engineers were moving much faster with AI, while others had little visibility into how those workflows actually worked.
That gap is what led me to build athrd.
athrd automatically attaches AI coding sessions to commits and surfaces them in pull requests, so reviewers can understand not just what changed, but why it changed. Instead of reviewing a diff in isolation, teams get the context, tradeoffs, and decision trail behind the code.
It also turns strong AI-assisted workflows into shared team knowledge. You can learn from your best AI-native engineers by seeing the prompts, iterations, and guidance behind the final result, not just the output.
athrd is built to fit into the workflow teams already use, and helps make AI-assisted development more reviewable, teachable, and trustworthy.
I’d love your feedback, especially if your team is figuring out how to review and scale AI-assisted code.
If you want, I can also make it: 1. more founder-personal, 2. more technical, 3. or shorter and punchier for Product Hunt.
Updated:
Hi I’m Greg.
After being part of the recent layoffs at Block, I kept thinking about something I saw firsthand: AI adoption across engineering teams was incredibly uneven. Some engineers were moving much faster with AI, while others had little visibility into how those workflows actually worked.
That gap is what led me to build athrd.
athrd automatically attaches AI coding sessions to commits and surfaces them in pull requests, so reviewers can understand not just what changed, but why it changed. Instead of reviewing a diff in isolation, teams get the context, tradeoffs, and decision trail behind the code.
It also turns strong AI-assisted workflows into shared team knowledge. You can learn from your best AI-native engineers by seeing the prompts, iterations, and guidance behind the final result, not just the output.
athrd is built to fit into the workflow teams already use, and helps make AI-assisted development more reviewable, teachable, and trustworthy.
If you're currently using Codex or Claude Code, I'd love your feedback.
Not sure if useful. But i saw the opportunity to build athr while at Block, not after. I attended so many demos and calls where people were sharing how their were using AI, skills, mcp, ect Everything was throuh demo recordings or a live screensharing. It was literaly impossible to actually see what the person did and how they actually used ai, codex, claude, or whatever else.
The Block context is useful. The layoff detail is probably not the strongest opening for a public Product Hunt comment. The sharper story is: you saw the problem while you were at Block, and the gap was that teams could watch AI demos but couldn’t inspect the actual workflow.
I’d rewrite it like this:
Hi, I’m Greg.
I started building athrd after seeing a clear gap while I was at Block: AI adoption across engineering teams was highly uneven, and the people getting the most out of tools like Codex and Claude Code were often the hardest to learn from.
I sat through a lot of demos, calls, recordings, and screen shares where engineers shared impressive AI workflows. But most of the time, you could only see the outcome. You couldn’t really inspect what they typed, how they guided the agent, what they tried, what failed, or why they made certain decisions.
That felt like a missing layer in the development workflow.
athrd automatically attaches AI coding sessions to commits and surfaces them in pull requests, so reviewers can understand not just what changed, but why it changed. Instead of reviewing a diff in isolation, teams can see the context, tradeoffs, and decision trail behind the code.
It also turns strong AI workflows into shared team knowledge. The best AI-native engineers stop being black boxes, and their process becomes something the rest of the team can actually learn from.
athrd fits into the workflow teams already use and helps make AI-assisted development more reviewable, teachable, and trustworthy.
If you’re using Codex or Claude Code, I’d love your feedback.
If you want, I can also make this more: 1. founder-personal, 2. sharper / more technical, 3. or more Product Hunt-y and concise.
I think saying I was laidoff from the very beginning is a differentiator. The resst looks great.
Then keep it, but make it work for you in the first sentence and pivot immediately to the insight. This version does that:
Hi, I’m Greg.
I was recently laid off from Block, and one of the biggest ideas I left with came from what I saw there: AI adoption across engineering teams was incredibly uneven, and the people getting the most out of tools like Codex and Claude Code were often the hardest to learn from.
I sat through a lot of demos, calls, recordings, and screen shares where engineers shared impressive AI workflows. But most of the time, you could only see the outcome. You couldn’t really inspect what they typed, how they guided the agent, what they tried, what failed, or why they made certain decisions.
That gap is what led me to build athrd.
athrd automatically attaches AI coding sessions to commits and surfaces them in pull requests, so reviewers can understand not just what changed, but why it changed. Instead of reviewing a diff in isolation, teams can see the context, tradeoffs, and decision trail behind the code.
It also turns strong AI workflows into shared team knowledge. The best AI-native engineers stop being black boxes, and their process becomes something the rest of the team can actually learn from.
athrd fits into the workflow teams already use and helps make AI-assisted development more reviewable, teachable, and trustworthy.
If you’re using Codex or Claude Code, I’d love your feedback.
Small note: I was recently laid off from Block is stronger and cleaner than I was part of the recent layoffs at Block.
If you want, I can also make this 15-20% shorter for better Product Hunt readability.