Edd Mann
@eddmann.com
the workflow is simple:
- global hotkey (option-shift-space)
- press space to record
- transcription auto-copies and can auto-paste
- recent history stored locally in swiftdata
- global hotkey (option-shift-space)
- press space to record
- transcription auto-copies and can auto-paste
- recent history stored locally in swiftdata
November 12, 2025 at 9:37 AM
the workflow is simple:
- global hotkey (option-shift-space)
- press space to record
- transcription auto-copies and can auto-paste
- recent history stored locally in swiftdata
- global hotkey (option-shift-space)
- press space to record
- transcription auto-copies and can auto-paste
- recent history stored locally in swiftdata
the core is pretty straightforward:
- whisperkit for transcription (base/small/medium models)
- mlx for ai post-processing (qwen 2.5, llama 3.2, phi-3.5)
- all models downloaded from hugging face
- all inference happens locally on your mac
- whisperkit for transcription (base/small/medium models)
- mlx for ai post-processing (qwen 2.5, llama 3.2, phi-3.5)
- all models downloaded from hugging face
- all inference happens locally on your mac
November 12, 2025 at 9:36 AM
the core is pretty straightforward:
- whisperkit for transcription (base/small/medium models)
- mlx for ai post-processing (qwen 2.5, llama 3.2, phi-3.5)
- all models downloaded from hugging face
- all inference happens locally on your mac
- whisperkit for transcription (base/small/medium models)
- mlx for ai post-processing (qwen 2.5, llama 3.2, phi-3.5)
- all models downloaded from hugging face
- all inference happens locally on your mac
i built a voice transcription app for macOS that runs 100% locally on your mac - no cloud, no api calls, complete privacy. 🎙️
here's how it came together 👇
here's how it came together 👇
November 12, 2025 at 9:36 AM
i built a voice transcription app for macOS that runs 100% locally on your mac - no cloud, no api calls, complete privacy. 🎙️
here's how it came together 👇
here's how it came together 👇
by monday morning, i pulled the repo from github… and ran pokemon red in my terminal. 🤯
November 11, 2025 at 1:19 PM
by monday morning, i pulled the repo from github… and ran pokemon red in my terminal. 🤯
friday night, i put together a PLAN md with cGPT’s help, then got the remote agent to start building step-by-step - each stage had clear goals and outcomes before moving on.
November 11, 2025 at 1:18 PM
friday night, i put together a PLAN md with cGPT’s help, then got the remote agent to start building step-by-step - each stage had clear goals and outcomes before moving on.
first step - come up with a name.
then, obviously, get cGPT to generate some artwork (because why not).
then, obviously, get cGPT to generate some artwork (because why not).
November 11, 2025 at 1:18 PM
first step - come up with a name.
then, obviously, get cGPT to generate some artwork (because why not).
then, obviously, get cGPT to generate some artwork (because why not).
The BEAM is more than a runtime - it’s #Erlang’s real superpower.
@shawnmc.cool explains how the BEAM isn’t just a VM, but an operating system for concurrency. Every process runs in complete isolation, enabling massive fault tolerance and reliability.
🎧 compiledconversations.com/12/%EF%BF%BC
@shawnmc.cool explains how the BEAM isn’t just a VM, but an operating system for concurrency. Every process runs in complete isolation, enabling massive fault tolerance and reliability.
🎧 compiledconversations.com/12/%EF%BF%BC
November 5, 2025 at 8:31 PM
The BEAM is more than a runtime - it’s #Erlang’s real superpower.
@shawnmc.cool explains how the BEAM isn’t just a VM, but an operating system for concurrency. Every process runs in complete isolation, enabling massive fault tolerance and reliability.
🎧 compiledconversations.com/12/%EF%BF%BC
@shawnmc.cool explains how the BEAM isn’t just a VM, but an operating system for concurrency. Every process runs in complete isolation, enabling massive fault tolerance and reliability.
🎧 compiledconversations.com/12/%EF%BF%BC
Concurrency without the chaos. ⚡️
@shawnmc.cool unpacks the power of OTP - #Erlang’s “standard library for concurrency.”
Learn how it turns thousands of lightweight processes into reliable, self-healing systems that just work.
🎧 Listen now: compiledconversations.com/12/%EF%BF%BC
@shawnmc.cool unpacks the power of OTP - #Erlang’s “standard library for concurrency.”
Learn how it turns thousands of lightweight processes into reliable, self-healing systems that just work.
🎧 Listen now: compiledconversations.com/12/%EF%BF%BC
November 4, 2025 at 8:31 PM
Concurrency without the chaos. ⚡️
@shawnmc.cool unpacks the power of OTP - #Erlang’s “standard library for concurrency.”
Learn how it turns thousands of lightweight processes into reliable, self-healing systems that just work.
🎧 Listen now: compiledconversations.com/12/%EF%BF%BC
@shawnmc.cool unpacks the power of OTP - #Erlang’s “standard library for concurrency.”
Learn how it turns thousands of lightweight processes into reliable, self-healing systems that just work.
🎧 Listen now: compiledconversations.com/12/%EF%BF%BC
Failure isn’t the enemy - it’s part of the design. ⚙️
In a recent episode of Compiled Conversations, @shawnmc.cool broke down Erlang’s “let it crash” philosophy - why letting things fail can actually make your systems more reliable, not less.
🎧 Listen now: compiledconversations.com/12/
In a recent episode of Compiled Conversations, @shawnmc.cool broke down Erlang’s “let it crash” philosophy - why letting things fail can actually make your systems more reliable, not less.
🎧 Listen now: compiledconversations.com/12/
November 3, 2025 at 6:52 PM
Failure isn’t the enemy - it’s part of the design. ⚙️
In a recent episode of Compiled Conversations, @shawnmc.cool broke down Erlang’s “let it crash” philosophy - why letting things fail can actually make your systems more reliable, not less.
🎧 Listen now: compiledconversations.com/12/
In a recent episode of Compiled Conversations, @shawnmc.cool broke down Erlang’s “let it crash” philosophy - why letting things fail can actually make your systems more reliable, not less.
🎧 Listen now: compiledconversations.com/12/
Context is everything when working with LLMs.
In episode 7 of Compiled Conversations, @ashleyhindle.com explains why managing context windows is critical to getting quality, reliable output.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify.
In episode 7 of Compiled Conversations, @ashleyhindle.com explains why managing context windows is critical to getting quality, reliable output.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify.
October 7, 2025 at 8:45 AM
Context is everything when working with LLMs.
In episode 7 of Compiled Conversations, @ashleyhindle.com explains why managing context windows is critical to getting quality, reliable output.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify.
In episode 7 of Compiled Conversations, @ashleyhindle.com explains why managing context windows is critical to getting quality, reliable output.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify.
Zero-shot, one-shot, few-shot - what do they really mean when prompting large language models?
In episode 7 of Compiled Conversations, @ashleyhindle.com breaks down each one.
🎧 Listen now: compiledconversations.com/7/
👉 Don’t miss future episodes - subscribe on Apple Podcasts or Spotify.
In episode 7 of Compiled Conversations, @ashleyhindle.com breaks down each one.
🎧 Listen now: compiledconversations.com/7/
👉 Don’t miss future episodes - subscribe on Apple Podcasts or Spotify.
October 5, 2025 at 6:18 PM
Zero-shot, one-shot, few-shot - what do they really mean when prompting large language models?
In episode 7 of Compiled Conversations, @ashleyhindle.com breaks down each one.
🎧 Listen now: compiledconversations.com/7/
👉 Don’t miss future episodes - subscribe on Apple Podcasts or Spotify.
In episode 7 of Compiled Conversations, @ashleyhindle.com breaks down each one.
🎧 Listen now: compiledconversations.com/7/
👉 Don’t miss future episodes - subscribe on Apple Podcasts or Spotify.
LLM agents get talked about a lot - but what are they, really?
In Episode 7 of Compiled Conversations, @ashleyhindle.com strips away the hype and explains how agents are, at their core, just a simple loop.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify
In Episode 7 of Compiled Conversations, @ashleyhindle.com strips away the hype and explains how agents are, at their core, just a simple loop.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify
October 4, 2025 at 6:11 PM
LLM agents get talked about a lot - but what are they, really?
In Episode 7 of Compiled Conversations, @ashleyhindle.com strips away the hype and explains how agents are, at their core, just a simple loop.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify
In Episode 7 of Compiled Conversations, @ashleyhindle.com strips away the hype and explains how agents are, at their core, just a simple loop.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify
In Functional Programming, functions aren't just helpers - they are the building blocks.
In the latest episode, @scottwlaschin.bsky.social breaks down how FP uses functions to compose behaviour, control flow, and model systems in a different way from OOP.
🎧 compiledconversations.com/9/
In the latest episode, @scottwlaschin.bsky.social breaks down how FP uses functions to compose behaviour, control flow, and model systems in a different way from OOP.
🎧 compiledconversations.com/9/
October 3, 2025 at 6:26 PM
In Functional Programming, functions aren't just helpers - they are the building blocks.
In the latest episode, @scottwlaschin.bsky.social breaks down how FP uses functions to compose behaviour, control flow, and model systems in a different way from OOP.
🎧 compiledconversations.com/9/
In the latest episode, @scottwlaschin.bsky.social breaks down how FP uses functions to compose behaviour, control flow, and model systems in a different way from OOP.
🎧 compiledconversations.com/9/
Exploring #MCP and @laravel.com MCP by building a @plex.tv integration 🎬
AI assistants now understand where I am in a film through subtitle analysis.
'What have I missed?' gets context-aware answers without spoilers.
Write up: eddmann.com/posts/contex...
Repo: github.com/eddmann/plex...
AI assistants now understand where I am in a film through subtitle analysis.
'What have I missed?' gets context-aware answers without spoilers.
Write up: eddmann.com/posts/contex...
Repo: github.com/eddmann/plex...
October 3, 2025 at 4:23 PM
Exploring #MCP and @laravel.com MCP by building a @plex.tv integration 🎬
AI assistants now understand where I am in a film through subtitle analysis.
'What have I missed?' gets context-aware answers without spoilers.
Write up: eddmann.com/posts/contex...
Repo: github.com/eddmann/plex...
AI assistants now understand where I am in a film through subtitle analysis.
'What have I missed?' gets context-aware answers without spoilers.
Write up: eddmann.com/posts/contex...
Repo: github.com/eddmann/plex...
Laravel Boost makes AI-assisted development faster and more reliable. ⚡
@ashleyhindle.com explains how it provides tested guidelines, its own MCP server, and version-specific docs for Laravel projects.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify for more.
@ashleyhindle.com explains how it provides tested guidelines, its own MCP server, and version-specific docs for Laravel projects.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify for more.
October 2, 2025 at 6:03 PM
Laravel Boost makes AI-assisted development faster and more reliable. ⚡
@ashleyhindle.com explains how it provides tested guidelines, its own MCP server, and version-specific docs for Laravel projects.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify for more.
@ashleyhindle.com explains how it provides tested guidelines, its own MCP server, and version-specific docs for Laravel projects.
🎧 Listen now: compiledconversations.com/7/
👉 Subscribe on Apple Podcasts or Spotify for more.
Error handling often clutters code.
Fortunately, functional programming offers a more elegant way.
@scottwlaschin.bsky.social explains the 'two-track' model 🚂 (railway-oriented programming) - keeping the success path clean while ensuring errors are always handled.
🎧 compiledconversations.com/9/
Fortunately, functional programming offers a more elegant way.
@scottwlaschin.bsky.social explains the 'two-track' model 🚂 (railway-oriented programming) - keeping the success path clean while ensuring errors are always handled.
🎧 compiledconversations.com/9/
September 29, 2025 at 6:26 PM
Error handling often clutters code.
Fortunately, functional programming offers a more elegant way.
@scottwlaschin.bsky.social explains the 'two-track' model 🚂 (railway-oriented programming) - keeping the success path clean while ensuring errors are always handled.
🎧 compiledconversations.com/9/
Fortunately, functional programming offers a more elegant way.
@scottwlaschin.bsky.social explains the 'two-track' model 🚂 (railway-oriented programming) - keeping the success path clean while ensuring errors are always handled.
🎧 compiledconversations.com/9/
Programming isn't one-size-fits-all.
In Episode 9 of Compiled Conversations, @scottwlaschin.bsky.social makes the case for being multi-paradigm - using the right language for the right job. Just like a tradesperson has more than one tool, a programmer should too.
🎧 compiledconversations.com/9/
In Episode 9 of Compiled Conversations, @scottwlaschin.bsky.social makes the case for being multi-paradigm - using the right language for the right job. Just like a tradesperson has more than one tool, a programmer should too.
🎧 compiledconversations.com/9/
September 28, 2025 at 6:26 PM
Programming isn't one-size-fits-all.
In Episode 9 of Compiled Conversations, @scottwlaschin.bsky.social makes the case for being multi-paradigm - using the right language for the right job. Just like a tradesperson has more than one tool, a programmer should too.
🎧 compiledconversations.com/9/
In Episode 9 of Compiled Conversations, @scottwlaschin.bsky.social makes the case for being multi-paradigm - using the right language for the right job. Just like a tradesperson has more than one tool, a programmer should too.
🎧 compiledconversations.com/9/
The Reverse Conway Maneuver flips the script on Conway’s Law.
In Episode 8 of Compiled Conversations, Scott explains how aligning teams with domains helps avoid conflicts - and why the number of teams you have inevitably shapes your system’s architecture.
🎧 Listen now: compiledconversations.com/8/
In Episode 8 of Compiled Conversations, Scott explains how aligning teams with domains helps avoid conflicts - and why the number of teams you have inevitably shapes your system’s architecture.
🎧 Listen now: compiledconversations.com/8/
September 27, 2025 at 8:27 PM
The Reverse Conway Maneuver flips the script on Conway’s Law.
In Episode 8 of Compiled Conversations, Scott explains how aligning teams with domains helps avoid conflicts - and why the number of teams you have inevitably shapes your system’s architecture.
🎧 Listen now: compiledconversations.com/8/
In Episode 8 of Compiled Conversations, Scott explains how aligning teams with domains helps avoid conflicts - and why the number of teams you have inevitably shapes your system’s architecture.
🎧 Listen now: compiledconversations.com/8/
Garbage in, garbage out. 🗑️
In Episode 8 of Compiled Conversations, Scott explains why good design starts with good inputs - and why even the best code can’t fix bad ones.
🎧 Listen now: compiledconversations.com/8/
👉 Subscribe on Apple Podcasts or Spotify so you don’t miss future episodes.
In Episode 8 of Compiled Conversations, Scott explains why good design starts with good inputs - and why even the best code can’t fix bad ones.
🎧 Listen now: compiledconversations.com/8/
👉 Subscribe on Apple Podcasts or Spotify so you don’t miss future episodes.
September 25, 2025 at 6:26 PM
Garbage in, garbage out. 🗑️
In Episode 8 of Compiled Conversations, Scott explains why good design starts with good inputs - and why even the best code can’t fix bad ones.
🎧 Listen now: compiledconversations.com/8/
👉 Subscribe on Apple Podcasts or Spotify so you don’t miss future episodes.
In Episode 8 of Compiled Conversations, Scott explains why good design starts with good inputs - and why even the best code can’t fix bad ones.
🎧 Listen now: compiledconversations.com/8/
👉 Subscribe on Apple Podcasts or Spotify so you don’t miss future episodes.
Writing code isn't just for computers - it's for people too.
@scottwlaschin.bsky.social explains how #DDD encourages developers to use the same language as non-developers; making code a tool for communication, not just execution.
🎧 compiledconversations.com/8/
@scottwlaschin.bsky.social explains how #DDD encourages developers to use the same language as non-developers; making code a tool for communication, not just execution.
🎧 compiledconversations.com/8/
September 23, 2025 at 5:11 PM
Writing code isn't just for computers - it's for people too.
@scottwlaschin.bsky.social explains how #DDD encourages developers to use the same language as non-developers; making code a tool for communication, not just execution.
🎧 compiledconversations.com/8/
@scottwlaschin.bsky.social explains how #DDD encourages developers to use the same language as non-developers; making code a tool for communication, not just execution.
🎧 compiledconversations.com/8/
🧝✨ Workshop update from the North Pole: Jolly JVM just delivered a #Java impl. of elf-lang featuring #Clojure-style immutable, persistent collections w/ structural sharing!
🔧 Impl: eddmann.com/santa-lang-w...
📖 Writeup: eddmann.com/posts/santa-...
🔧 Impl: eddmann.com/santa-lang-w...
📖 Writeup: eddmann.com/posts/santa-...
September 16, 2025 at 3:31 PM
🧝✨ Workshop update from the North Pole: Jolly JVM just delivered a #Java impl. of elf-lang featuring #Clojure-style immutable, persistent collections w/ structural sharing!
🔧 Impl: eddmann.com/santa-lang-w...
📖 Writeup: eddmann.com/posts/santa-...
🔧 Impl: eddmann.com/santa-lang-w...
📖 Writeup: eddmann.com/posts/santa-...
I may have taken the Christmas-themed programming language thing a bit too far...
Introducing the santa-lang Workshop: AI agents (elves) implementing elf-lang across different programming languages using shared test suites and stage-gated development.
📖 eddmann.com/posts/santa-...
#AI #LLM
Introducing the santa-lang Workshop: AI agents (elves) implementing elf-lang across different programming languages using shared test suites and stage-gated development.
📖 eddmann.com/posts/santa-...
#AI #LLM
September 15, 2025 at 4:33 PM
I may have taken the Christmas-themed programming language thing a bit too far...
Introducing the santa-lang Workshop: AI agents (elves) implementing elf-lang across different programming languages using shared test suites and stage-gated development.
📖 eddmann.com/posts/santa-...
#AI #LLM
Introducing the santa-lang Workshop: AI agents (elves) implementing elf-lang across different programming languages using shared test suites and stage-gated development.
📖 eddmann.com/posts/santa-...
#AI #LLM
🎄 Getting into the Christmas spirit a little early!
I set a team of #LLM 'elves' (agents) the task of re-implementing a subset of a language I designed several years ago - introducing... elf-lang.
🧵👇
I set a team of #LLM 'elves' (agents) the task of re-implementing a subset of a language I designed several years ago - introducing... elf-lang.
🧵👇
September 9, 2025 at 3:30 PM
🎄 Getting into the Christmas spirit a little early!
I set a team of #LLM 'elves' (agents) the task of re-implementing a subset of a language I designed several years ago - introducing... elf-lang.
🧵👇
I set a team of #LLM 'elves' (agents) the task of re-implementing a subset of a language I designed several years ago - introducing... elf-lang.
🧵👇
Strategic vs Tactical #DDD - know the difference?
Strategic: define domains, goals & guide architecture
Tactical: patterns that turn strategy into code
One devs ❤️. The other drives impact. 💪
🎧 compiledconversations.com/4/ compiledconversations.com/5/
👉 Subscribe on Apple Podcasts or Spotify
Strategic: define domains, goals & guide architecture
Tactical: patterns that turn strategy into code
One devs ❤️. The other drives impact. 💪
🎧 compiledconversations.com/4/ compiledconversations.com/5/
👉 Subscribe on Apple Podcasts or Spotify
September 2, 2025 at 4:57 PM
Strategic vs Tactical #DDD - know the difference?
Strategic: define domains, goals & guide architecture
Tactical: patterns that turn strategy into code
One devs ❤️. The other drives impact. 💪
🎧 compiledconversations.com/4/ compiledconversations.com/5/
👉 Subscribe on Apple Podcasts or Spotify
Strategic: define domains, goals & guide architecture
Tactical: patterns that turn strategy into code
One devs ❤️. The other drives impact. 💪
🎧 compiledconversations.com/4/ compiledconversations.com/5/
👉 Subscribe on Apple Podcasts or Spotify