William (Bill) Kennedy
banner
goinggo.net
William (Bill) Kennedy
@goinggo.net
⌯Go ⦁ Walking the line between correctness and comprehension ⦁ bill@ardanlabs.com ⦁ wife(aleintech.com) ⦁ GoBridge ⦁ DJ Name (Low Key Yay)
Pinned
Here are my 2 #golang talks from @GopherCon

Understanding Escape Analysis to Speed Up Your Code
www.youtube.com/watch?v=FUm0...

Go’s Trace Tooling and Concurrency
www.youtube.com/watch?v=Gqo0...
Reposted by William (Bill) Kennedy
🇧🇷 São Paulo devs — hear it from @goinggo.net himself!

Ardan Labs Connect 2026 is coming March 4th–5th.

🛠 Day 1: Ultimate Software Design
🤖 Day 2: Ultimate AI
🤝 + Golang SP community meetup

🎟️ www.ardanlabs.com/events/ardan...

Presented by Ardan Labs x GopherCon Latam
February 11, 2026 at 5:23 PM
Why has it taken 56 years for me to discover this spoon! I just bought 6! 😂
February 7, 2026 at 4:49 PM
My lighting talk at FOSDEM in the #golang room last weekend.

Introducing Kronk

cuddly.tube/w/mBtDbXQLde...
Go Lightning talks @ FOSDEM 2026
Talks: Kronk: hardware accelerated local inference by William Kennedy Re-imagining the Tamagotchi with Go by Andy Williams A Go app on mobile, the easy way by Tanguy ⧓ Herrmann Write a microservice...
cuddly.tube
February 7, 2026 at 2:38 PM
Reposted by William (Bill) Kennedy
Our YouTube short of the week is here!💻

This week revolves around “Increasing ROI with AI” from @goinggo.net ’s podcast episode with featured guest Peter Swimm! 🤓

🎥: youtube.com/shorts/Cdx96...

#ArdanLabsPodcast #Shorts #YouTube #SoftwareEngineering
Increasing ROI with AI
YouTube video by Ardan Labs
youtube.com
February 5, 2026 at 6:54 PM
Kronk v1.17.7 is close to feature complete for all the things I wanted to accomplish.

It's ready for people to start building software against. There is a manual at the root of the project. #golang
February 7, 2026 at 1:18 PM
Reposted by William (Bill) Kennedy
Podcast Ep w/ @aleintech.com & Oscar Hedaya is out!

In this episode, Oscar shares his journey from New Jersey to Miami, the childhood financial challenges that shaped his work ethic, & the lessons learned from college, job searching, & early setbacks.

🎥Watch: youtu.be/E5Zp5yII69M
Hardware, Innovation, and The Space Safe with Oscar Hedaya
YouTube video by Ardan Labs
youtu.be
February 4, 2026 at 4:16 PM
Reposted by William (Bill) Kennedy
🇧🇷Calling our Brazil devs!

Mark March 4th & 5th on your calendar for two hands-on days with @goinggo.net
🛠 Day 1: Ult. Software Design
🤖 Day 2: Ultimate AI
🤝 Bonus: Golang SP community meetup

🔗 www.ardanlabs.com/events/ardan...

In partnership with Gophercon Latam
February 4, 2026 at 5:30 PM
Reposted by William (Bill) Kennedy
Now my main man @goinggo.net speaking at the AI Plumbers #fosdem unconf about Kronk/yzma

#golang #ml #llama
February 2, 2026 at 10:37 AM
Reposted by William (Bill) Kennedy
Ult. Software Design LIVE Schedule📅

Join @goinggo.net & @kenriquezcodes.bsky.social in this week's stream:
Thu. 2/5 from 11am-1pm EST

Stay tuned 1 hour before the LIVE show for the stream link!😎

📽️Rewatch the last 69 EPs here: bit.ly/3CShDOS
#golang #SoftwareDesignLive
February 2, 2026 at 6:30 PM
Reposted by William (Bill) Kennedy
Everybody asking for ML-DSA in Go:
January 29, 2026 at 7:05 PM
Reposted by William (Bill) Kennedy
🇧🇷 Brazil devs — São Paulo in March?
Join us for Ardan Labs Connect 2026 on March 4th & 5th

Spend two intensive days learning directly from @goinggo.net :

🧠 Day 1: Ultimate Software Design with Go
🤖 Day 2: Ultimate AI

👉 Get the details and reserve your spot:
www.ardanlabs.com/events/ardan...
January 29, 2026 at 5:03 PM
Kronk v1.16.3 is the best version yet. Lot's of improvements, including the gpt-oss model is working better with tool calling. Right now it's my best model with OpenCode. The Qwen3 is my best model with Cline.

Nice improvements with the BUI and the core performance.

github.com/ardanlabs/kr...
GitHub - ardanlabs/kronk: This project lets you use Go for hardware accelerated local inference with llama.cpp directly integrated into your applications via the yzma module. Kronk provides a high-lev...
This project lets you use Go for hardware accelerated local inference with llama.cpp directly integrated into your applications via the yzma module. Kronk provides a high-level API that feels simil...
github.com
January 29, 2026 at 3:58 PM
Reposted by William (Bill) Kennedy
January 29, 2026 at 8:03 AM
Reposted by William (Bill) Kennedy
Gophers, 1-hour countdown for stream #69! 😎
Join @goinggo.net & @kenriquezcodes.bsky.social as they go live at 11 am EST!

📽️ Today's stream link: youtube.com/live/004IdO6...
#golang #Livestream #SoftwareDesign
Ultimate Go Software Design LIVE: Ep.69
YouTube video by Ardan Labs
youtube.com
January 27, 2026 at 3:00 PM
Spent more time today putting Claude Code head to head with AMP.

AMP is just more efficient and writes better code. It just does every time.

I'm at a point where I'm only using Claude to check AMP's work and review code to see if it can find anything AMP misses. That's rare.
January 26, 2026 at 10:04 PM
Reposted by William (Bill) Kennedy
Ult. Software Design LIVE Schedule📅
Join @goinggo.net & @kenriquezcodes.bsky.social in this week's streams:
Tue. 1/27 from 11am-1pm EST

Stay tuned 1 hour before the LIVE show for the stream link!😎

📽️Rewatch the last 68 EPs here: bit.ly/3CShDOS
#golang #SoftwareDesignLive
January 26, 2026 at 7:45 PM
Version 1.15.5 has been released of #Kronk. Best version yet. Performance improvements and working with Cline, Claude, and OpenCode.

This model has worked great on my M4 128Gig

Qwen3-Coder-30B-A3B-Instruct-UD-Q8_K_XL

github.com/ardanlabs/kr...
GitHub - ardanlabs/kronk: This project lets you use Go for hardware accelerated local inference with llama.cpp directly integrated into your applications via the yzma module. Kronk provides a high-lev...
This project lets you use Go for hardware accelerated local inference with llama.cpp directly integrated into your applications via the yzma module. Kronk provides a high-level API that feels simil...
github.com
January 25, 2026 at 6:38 PM
Taking back what I said. Finding that if I have #AMP check #Claude and visa versa, I find things the other doesn't.

@aleintech.com said, "It takes a team" 🤯

I have to think of these different Agents as team members who can feed of each other!

bsky.app/profile/goin...
I can safely say that #AMP is better than #ClaudeCode

Asked:
- Both to plan a complex change.
- Claude to implement it.
- AMP to review it and it found problems.
- AMP to fix the code.
- Claude to review it.
January 25, 2026 at 5:21 PM
I can safely say that #AMP is better than #ClaudeCode

Asked:
- Both to plan a complex change.
- Claude to implement it.
- AMP to review it and it found problems.
- AMP to fix the code.
- Claude to review it.
January 24, 2026 at 8:03 PM
Excited with `GLM-4.7-Flash-UD-Q8_K_XL`. It's working great with Cline on my M4. Works solid with Claude Code as well, but Cline is faster using the model.

Of course I'm using Kronk as the model server.

This model will allow me to use an Agent when disconnected!!

huggingface.co/unsloth/GLM-...
unsloth/GLM-4.7-Flash-GGUF · Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
January 24, 2026 at 12:51 AM
I will be going to #FOSDEM and giving a talk about #kronk at the Fringe AI Plumbers (un)conference event. #golang

If you will be in town on Monday 2nd, come hang out!

luma.com/3nc5qaje
FOSDEM’26 fringe: AI Plumbers (un)conference · Luma
We got lucky and are running AI Devroom at FOSDEM26 again. Of course, we would also love to run a 2nd installment of the AI Plumbers Fringe Event! This time it…
luma.com
January 22, 2026 at 10:39 PM
Kronk 1.14.7 is the best version yet. #golang
* Support for Cline
* System prompt caching
* Improved tool call parsing
* Parallel batch processing
* Lot's of bug fixes

github.com/ardanlabs/kr...
GitHub - ardanlabs/kronk: This project lets you use Go for hardware accelerated local inference with llama.cpp directly integrated into your applications via the yzma module. Kronk provides a high-lev...
This project lets you use Go for hardware accelerated local inference with llama.cpp directly integrated into your applications via the yzma module. Kronk provides a high-level API that feels simil...
github.com
January 22, 2026 at 10:35 PM
Reposted by William (Bill) Kennedy
🇧🇷 Heading to Brazil in March?
Make sure Ardan Labs Connect 2026 is on your calendar.
📍 São Paulo | 🗓 March 4th–5th

Two immersive days of hands-on training with @goinggo.net :
Day 1: Ultimate Software Design with Go
Day 2: Ultimate AI Application Development

🔗 www.ardanlabs.com/events/ardan...
January 22, 2026 at 5:45 PM
Reposted by William (Bill) Kennedy
Our YouTube short of the week is here!💻

This week revolves around “Understanding Agent Performance & Adaptive Loops” from @goinggo.net's podcast episode with featured guest Kenneth Stott! 🤓

🎥 : youtube.com/shorts/hDOcc...

#ArdanLabsPodcast #Shorts #YouTube #SoftwareEngineering #LLM
Understanding Agent Performance and Adaptive Loops
YouTube video by Ardan Labs
youtube.com
January 22, 2026 at 8:53 PM