Mick.net - SaaS, AI agents, Aviation
@mick-net.bsky.social
Posting about my (remote) SaaS journeys, AI agents, Aviation, and other nerdy topics.
Maker of BestTime.app - building Aviation.bot
Maker of BestTime.app - building Aviation.bot
Reduce eye strain using this free Mac app using eye-tracking.
It fades your screen until you blink again to encourage you to blink more.
It fades your screen until you blink again to encourage you to blink more.
March 27, 2025 at 11:25 AM
Reduce eye strain using this free Mac app using eye-tracking.
It fades your screen until you blink again to encourage you to blink more.
It fades your screen until you blink again to encourage you to blink more.
I converted my whole repo into one big markdown file. Dumped it in Google Gemini's 1M token context, and asked me to help write cursor rules based on my existing core logic.
February 12, 2025 at 6:59 PM
I converted my whole repo into one big markdown file. Dumped it in Google Gemini's 1M token context, and asked me to help write cursor rules based on my existing core logic.
BestTime.app API has a steep learning curve, and it's pretty good at suggesting the right API when the user explain's their use-case.
It can also give a cost estimate and helps the user pick a plan, and helps to users navigate to the right page (it returns hyperlinks in the chat). 2/7
It can also give a cost estimate and helps the user pick a plan, and helps to users navigate to the right page (it returns hyperlinks in the chat). 2/7
February 11, 2025 at 2:02 AM
BestTime.app API has a steep learning curve, and it's pretty good at suggesting the right API when the user explain's their use-case.
It can also give a cost estimate and helps the user pick a plan, and helps to users navigate to the right page (it returns hyperlinks in the chat). 2/7
It can also give a cost estimate and helps the user pick a plan, and helps to users navigate to the right page (it returns hyperlinks in the chat). 2/7
Created a really good answering customer support bot 🤖 by inputting all my SaaS API docs, blogs, tutorials, website hyperlinks, pricing table & examples and common email request replies into Gemini's Flash 2.0 1M context window along with a nice system prompt. 🧵⬇️ 1/7
February 11, 2025 at 2:02 AM
Created a really good answering customer support bot 🤖 by inputting all my SaaS API docs, blogs, tutorials, website hyperlinks, pricing table & examples and common email request replies into Gemini's Flash 2.0 1M context window along with a nice system prompt. 🧵⬇️ 1/7
Try this when Sentry reports a bug 🐛:
- Create GitHub issue from within Sentry
- Paste GH issue link into Cursor compose and tell it to get the issue from GH
- Cursor uses GH MCP tool to see the issue 🤖
- Cursor digs through your code and suggests a fix 🦾
- Create GitHub issue from within Sentry
- Paste GH issue link into Cursor compose and tell it to get the issue from GH
- Cursor uses GH MCP tool to see the issue 🤖
- Cursor digs through your code and suggests a fix 🦾
February 10, 2025 at 2:14 AM
Try this when Sentry reports a bug 🐛:
- Create GitHub issue from within Sentry
- Paste GH issue link into Cursor compose and tell it to get the issue from GH
- Cursor uses GH MCP tool to see the issue 🤖
- Cursor digs through your code and suggests a fix 🦾
- Create GitHub issue from within Sentry
- Paste GH issue link into Cursor compose and tell it to get the issue from GH
- Cursor uses GH MCP tool to see the issue 🤖
- Cursor digs through your code and suggests a fix 🦾
Created an intro voice message for the AviationBot landing page.
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT.
The hero video is meant to convince users to sign up for the app.
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT.
The hero video is meant to convince users to sign up for the app.
February 7, 2025 at 4:52 AM
Created an intro voice message for the AviationBot landing page.
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT.
The hero video is meant to convince users to sign up for the app.
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT.
The hero video is meant to convince users to sign up for the app.
Created an intro voice message for the AviationBot landing page.
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT.
The hero video is meant to convince users to sign up for the app.
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT.
The hero video is meant to convince users to sign up for the app.
February 7, 2025 at 4:46 AM
Created an intro voice message for the AviationBot landing page.
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT.
The hero video is meant to convince users to sign up for the app.
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT.
The hero video is meant to convince users to sign up for the app.
Created an intro voice message for the AviationBot landing page.
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT. Does it sounds real enough? Maybe need to tweak the pauses and add a screen recording
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT. Does it sounds real enough? Maybe need to tweak the pauses and add a screen recording
February 7, 2025 at 4:38 AM
Created an intro voice message for the AviationBot landing page.
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT. Does it sounds real enough? Maybe need to tweak the pauses and add a screen recording
10 minutes of work by instructing ChatGPT to generate a 1-minute transcript based on the landing page and a tutorial + PlayHT. Does it sounds real enough? Maybe need to tweak the pauses and add a screen recording
Easily convert a GitHub repo to LLM-readable text with Uithub .com.
🪄 Simply change one character in the GitHub URL or folder to Uithub, and it works.
Select TXT and use this URL as documentation for your Cursor Docs library 🔗
Tada, now you have a smarter Cursor agent. 😇
🪄 Simply change one character in the GitHub URL or folder to Uithub, and it works.
Select TXT and use this URL as documentation for your Cursor Docs library 🔗
Tada, now you have a smarter Cursor agent. 😇
February 5, 2025 at 7:04 PM
Easily convert a GitHub repo to LLM-readable text with Uithub .com.
🪄 Simply change one character in the GitHub URL or folder to Uithub, and it works.
Select TXT and use this URL as documentation for your Cursor Docs library 🔗
Tada, now you have a smarter Cursor agent. 😇
🪄 Simply change one character in the GitHub URL or folder to Uithub, and it works.
Select TXT and use this URL as documentation for your Cursor Docs library 🔗
Tada, now you have a smarter Cursor agent. 😇
This illustrates how reasoning models have sped up the capability gain of AI.
(GPQA) is a series of multiple-choice problems that internet access doesn't help PhDs with access to the internet get 34% right on this test outside their specialty, 81% inside their specialty. This
(GPQA) is a series of multiple-choice problems that internet access doesn't help PhDs with access to the internet get 34% right on this test outside their specialty, 81% inside their specialty. This
February 4, 2025 at 7:00 PM
This illustrates how reasoning models have sped up the capability gain of AI.
(GPQA) is a series of multiple-choice problems that internet access doesn't help PhDs with access to the internet get 34% right on this test outside their specialty, 81% inside their specialty. This
(GPQA) is a series of multiple-choice problems that internet access doesn't help PhDs with access to the internet get 34% right on this test outside their specialty, 81% inside their specialty. This
@Cursor can now access my PostgreSQL database directly, through an MCP server! 🤯
Cursor's compose agents sometimes struggle to create one-shot good code as they are not aware of the data in the DB.
Hopefully I can improve that by making it DB context aware🧵⤵️
Cursor's compose agents sometimes struggle to create one-shot good code as they are not aware of the data in the DB.
Hopefully I can improve that by making it DB context aware🧵⤵️
February 3, 2025 at 2:33 AM
@Cursor can now access my PostgreSQL database directly, through an MCP server! 🤯
Cursor's compose agents sometimes struggle to create one-shot good code as they are not aware of the data in the DB.
Hopefully I can improve that by making it DB context aware🧵⤵️
Cursor's compose agents sometimes struggle to create one-shot good code as they are not aware of the data in the DB.
Hopefully I can improve that by making it DB context aware🧵⤵️
Better AI models aren't new, but they are surpassing benchmarks at an accelerating pace. With OpenAI's O3 (released today?), progress is advancing faster (See the last data points).
x.com/AndrewCurra...
x.com/AndrewCurra...
January 31, 2025 at 6:28 AM
Better AI models aren't new, but they are surpassing benchmarks at an accelerating pace. With OpenAI's O3 (released today?), progress is advancing faster (See the last data points).
x.com/AndrewCurra...
x.com/AndrewCurra...
Excited about upcoming releases in 2025. Meanwhile I guess this will still boosts Nvidia revenue. However, I'm curious how much CUDA still plays a roll at inference time vs training time.
January 30, 2025 at 4:03 AM
Excited about upcoming releases in 2025. Meanwhile I guess this will still boosts Nvidia revenue. However, I'm curious how much CUDA still plays a roll at inference time vs training time.
New reasoning systems can generate higher-quality training data than traditional synthetic data—improving with more active users. 💡
OpenAI benefits massively from this, as they have the most users & funding (thanks to Stargate), creating a compounding advantage. 🚀
OpenAI benefits massively from this, as they have the most users & funding (thanks to Stargate), creating a compounding advantage. 🚀
January 30, 2025 at 4:03 AM
New reasoning systems can generate higher-quality training data than traditional synthetic data—improving with more active users. 💡
OpenAI benefits massively from this, as they have the most users & funding (thanks to Stargate), creating a compounding advantage. 🚀
OpenAI benefits massively from this, as they have the most users & funding (thanks to Stargate), creating a compounding advantage. 🚀
Nvidia $NVDA stock dropped sharply, meanwhile ARC AGI's analysis of R1 also suggest the Javons Paradox where more efficient reasoning models result in:
- more $ spend for higher accuracy, reliability
- more $ spend on inference instead of $ training
🧵⬇️
- more $ spend for higher accuracy, reliability
- more $ spend on inference instead of $ training
🧵⬇️
January 30, 2025 at 4:03 AM
Nvidia $NVDA stock dropped sharply, meanwhile ARC AGI's analysis of R1 also suggest the Javons Paradox where more efficient reasoning models result in:
- more $ spend for higher accuracy, reliability
- more $ spend on inference instead of $ training
🧵⬇️
- more $ spend for higher accuracy, reliability
- more $ spend on inference instead of $ training
🧵⬇️
They claim to barely have GPU's. Meanwhile, AI labs with billions of funding and GPU power struggle to meet demand (although their models are probably not that efficient as V3/ R1).
January 28, 2025 at 2:40 AM
They claim to barely have GPU's. Meanwhile, AI labs with billions of funding and GPU power struggle to meet demand (although their models are probably not that efficient as V3/ R1).
I'm also skeptical about the DeepSeek hype. They may have developed a strong reasoning model, but it requires significant computational power at inference. How are they managing all those millions of requests 98% cheaper hardly any GPUs? Something seems off. 🤔
January 28, 2025 at 2:40 AM
I'm also skeptical about the DeepSeek hype. They may have developed a strong reasoning model, but it requires significant computational power at inference. How are they managing all those millions of requests 98% cheaper hardly any GPUs? Something seems off. 🤔
I tested the R1 distilled 32B model on my Mac, It solves flight engineering compliance questions other models like 4o could not. Unlike O1, which seems overconfident, R1 questions itself and follows output format instructions more closely, even though O1 also got answers right.
January 21, 2025 at 7:53 AM
I tested the R1 distilled 32B model on my Mac, It solves flight engineering compliance questions other models like 4o could not. Unlike O1, which seems overconfident, R1 questions itself and follows output format instructions more closely, even though O1 also got answers right.
Trying out ChatGPT Tasks Beta 👨💻. What will you use it for? 🤔
January 15, 2025 at 2:00 AM
Trying out ChatGPT Tasks Beta 👨💻. What will you use it for? 🤔
Can AI agents make live edits in (Word/Google) docs? 🤖
New AI code IDEs like Cursor and Windsurf can use an LLM to automatically figure out which lines to update and show a diff. It would be great if there were something similar for textual documents like Microsoft Word and Google Docs. 📝
New AI code IDEs like Cursor and Windsurf can use an LLM to automatically figure out which lines to update and show a diff. It would be great if there were something similar for textual documents like Microsoft Word and Google Docs. 📝
December 3, 2024 at 4:28 AM
Can AI agents make live edits in (Word/Google) docs? 🤖
New AI code IDEs like Cursor and Windsurf can use an LLM to automatically figure out which lines to update and show a diff. It would be great if there were something similar for textual documents like Microsoft Word and Google Docs. 📝
New AI code IDEs like Cursor and Windsurf can use an LLM to automatically figure out which lines to update and show a diff. It would be great if there were something similar for textual documents like Microsoft Word and Google Docs. 📝
🚀 Started outranking the official EU flight safety agency website thanks to custom snippets on some AviationBot pages. 🚀✈️
Any SEO experts know how can I further improve this? Snippets seem to be only working when the user is asking specific questions besides a reference.
Any SEO experts know how can I further improve this? Snippets seem to be only working when the user is asking specific questions besides a reference.
December 2, 2024 at 11:49 AM
🚀 Started outranking the official EU flight safety agency website thanks to custom snippets on some AviationBot pages. 🚀✈️
Any SEO experts know how can I further improve this? Snippets seem to be only working when the user is asking specific questions besides a reference.
Any SEO experts know how can I further improve this? Snippets seem to be only working when the user is asking specific questions besides a reference.
Cursor has now agents inside the IDE. What is your experience. How does it compare now to Windsurf?
November 25, 2024 at 1:22 AM
Cursor has now agents inside the IDE. What is your experience. How does it compare now to Windsurf?
Is anyone aware of a tool that takes all documentation (GitHub dir or sites) and outputs it to an LLM markdown format? 🤔
LLMs haven't caught up with LangGraph's features and APIs, so it would be great to feed all docs/examples into a long-context LLM/ Cursor docs.
@LangChainAI
LLMs haven't caught up with LangGraph's features and APIs, so it would be great to feed all docs/examples into a long-context LLM/ Cursor docs.
@LangChainAI
November 22, 2024 at 9:00 AM
Is anyone aware of a tool that takes all documentation (GitHub dir or sites) and outputs it to an LLM markdown format? 🤔
LLMs haven't caught up with LangGraph's features and APIs, so it would be great to feed all docs/examples into a long-context LLM/ Cursor docs.
@LangChainAI
LLMs haven't caught up with LangGraph's features and APIs, so it would be great to feed all docs/examples into a long-context LLM/ Cursor docs.
@LangChainAI
How do I fix this glitch, @cursor_ai?
I love using Cursor, but every time I copy and paste HTML snippets, it sneaks in an extra closing div on the first row! It's a stealthy disruptor that derails my workflow until I spot it.
I love using Cursor, but every time I copy and paste HTML snippets, it sneaks in an extra closing div on the first row! It's a stealthy disruptor that derails my workflow until I spot it.
November 22, 2024 at 1:26 AM
How do I fix this glitch, @cursor_ai?
I love using Cursor, but every time I copy and paste HTML snippets, it sneaks in an extra closing div on the first row! It's a stealthy disruptor that derails my workflow until I spot it.
I love using Cursor, but every time I copy and paste HTML snippets, it sneaks in an extra closing div on the first row! It's a stealthy disruptor that derails my workflow until I spot it.