LowKeyDevs retweeted
got my PR merged into googlecloudplatform/scion
added mocked cloud run lifecycle tests so the runtime can be tested without hitting gcp
github.com/GoogleCloudPlatfo…
genuinely forgot to post about this 😳
(my first merged pr btw)
LowKeyDevs retweeted
built SaaT (saat, turkish for time)
implemented 11 distributed clock algorithms from scratch in Go. zero frameworks.
from physical wallclocks to lamport clocks and vector clocks, covering google's truetime, interval tree clocks, and HLC. compared tradeoffs and how they resolve conflicts
also built a network simulator with jitter to audit causality and test concurrent edge cases across them
[ link : github.com/dexisback/Saat ]
LowKeyDevs retweeted
welcoming nuza v1.0.0!!
(the form i wanted nuza to take)
i started working on nuza because i needed a software that is simple, lightweight, has limited AI assistance and not bloated unlike other alternatives
was running out of space on my mac so i knew i had to create something of my own
the purpose was simple - apple notes but with cursor like editor that can assist me with grammar, phrasing and overall flow while i write blogs for my website
writing has always been my hobby so i needed a way to integrate AI into this space and thats how nuza came to be
LowKeyDevs retweeted
I started working at Commenda as an intern and enjoyed working with Spencer and Yaacov so much that I joined them full time. Working at a company where decisions are open sourced really teaches you a lot
It is easier to be curious about most ideas if you just dive a little bit deeper than the surface level.
I am looking for smart generalist interns who are great at absorbing ambition and want to produce some of their best work.
docs.google.com/document/d/1…
LowKeyDevs retweeted
today im launching gheppo.
gheppo is a beautiful terminal app built in go, lipgloss, bubble tea to show your github and leetcode contributions on every terminal session startup.
this is something i'd personally use to keep track and remind me of my green boxes whenever i slack off.
shipped with 5 configurable themes (github/catppuccin/nord/mono/gruvbox), works on all three: linux/macOS/windows
run a single curl command, one time setup. thats it.
[ link in comments ]
LowKeyDevs retweeted
Today, we received some great news for Gregale.
We’ve been trying to get Gregale accepted into the Neon Agent Program.
Since we’re still very early, Neon also granted us $500 in credits to help us build.
Seeing our idea get validated like this is incredibly exciting.
Back to building.
If you want to become a beta user of Gregale, DM me. It’ll be completely free during the beta.
LowKeyDevs retweeted
YOU WANT TO LEARN THIS IF YOU BUILD BACKENDS
How does scale to zero actually work?
It solves a simple but important infrastructure problem
How do you run zero instances when there’s no traffic, then handle the next request without booting the application from scratch?
The answer is:
Freeze the application. Save its state. Restore it when traffic comes back.
Here’s the basic flow:
Freeze the application. Save its state. Restore it when traffic comes back.
Here’s the basic flow:
No traffic → 0 active instances
Instead of destroying the environment, the system freezes it:
Running app
↓ freeze
Memory + CPU state → snapshot
↓ save
Disk
Now there’s no active instance using RAM.
When a request arrives:
Restore the snapshot
↓
Resume the application
↓
Handle the request
The application doesn’t have to start from scratch. It wakes up from where it left off.
That’s the core idea behind fast scale-to-zero:
Don’t restart the application.
Pause it.
Snapshot it.
Restore it when you need.
This is one reason microVMs are so interesting for serverless infrastructure.
We’re building this into Gregale.
We’ll be completely free during Beta.
gregale.dev
LowKeyDevs retweeted
TrenTorch.com is finally out.
The best way to learn Frontier ML for FREE
We cover things from
- Statistic for ML
- Classical ML
- Deep Learning
- Transformers & LLMS
- Distributed Training
- Reinforcement Learning
- Inference
ALL FOR FREEEEE
Feel free to check out the repo as well.
github.com/trentorch/trentor…
Amazing dev team
@Rocky_T07 @ShivtejG236 @aadityansha_06
Looking for sponsors
@vast_ai @modal @SarvamAI @hydra_db @googleaidevs @OpenAIDevs @OpenRouter @GroqLLC @ClaudeDevs @nebiusai @iitmadras @theo @ezSnippet @waitin4agi_ @kirat_tw @sheryians_ @ApnacollegeX @PhysicswallahAP
Everyone says learn inference, learn CUDA,learn ML
But almost nobody tells you exactly what to learn, how to learn it, or how to know whether you’ve actually learned it.
Introducing Trentorch.com a free, open-source platform to learn and practice Machine Learning, CUDA, and inference through hands-on coding.
Instead of just watching lectures or reading theory, you’ll:
• Learn the theory behind each concept
• Follow step-by-step implementation examples
• Solve coding challenges based on what you learned
• Implement algorithms from scratch, from foundational ML to Inference and CUDA kernel
• Practice problems in a Codeforces-style environment
• Get new Problem of the Day challenges with ratings
The goal isn't just to teach you how to write the code.
It’s to help you understand what your code is actually doing underneath.
We’re building Trentorch because advanced education shouldn’t be locked behind expensive subscriptions.
Money shouldn’t be the barrier to learning ML, inference, or CUDA.
A free alternative to platforms that charge students heavily every month.
signup and start solving problems now
Trentorch.com
LowKeyDevs retweeted
and now i am officially open to work
looking for backend engineering roles, specifically in Golang
i've spent the past year building projects and getting deeper into backend and systems development
projects below
stevenstank.vercel.app
github.com/stevenstank
LowKeyDevs retweeted
From tokenization → prefill → decode → KV cache → PagedAttention → continuous batching.
Still figuring out what really happens between a prompt and the next token.Documenting the journey as I learn.
LowKeyDevs retweeted
so i wanted to understand how containers actually work, so i built one from scratch in go
meet forge, my own small container runtime
it uses linux primitives for process isolation, filesystems, cgroups, networking, oci images, and the container lifecycle
heres a quick demo
LowKeyDevs retweeted
Introducing Sumora
You tell it who you want to sell to. It finds prospects on LinkedIn and Instagram, researches them, spots buying signals, and ranks who is worth contacting. Then it writes the outreach, follows up, and can build a highly personalized lead magnet for every prospect.
LowKeyDevs retweeted
New underleaf.xyz update!
New:
Drawing block with full freehand canvas drawing support
Mermaid diagram block
Inline AI on diagram, table, chart and a lot more blocks
and a lot more subtle but cool things
Would love your feedback
LowKeyDevs retweeted
notes on agents, control planes & the production gap.
(first video of me yapping, more coming soon ^^)
LowKeyDevs retweeted
Introducing underleaf.xyz
It’s a brand new document editing experience, built for the agent era from the ground up.
You use Notion-like blocks to structure your documents or you can just ask your agent to do things for you.
You and your agent work together.
LowKeyDevs retweeted
I have tried a lot of productivity applications and online personal assistants including claude, but nothing beats an application that watches every single keystroke and screen I see.
So I made Thunder. A MacOS watcher application.
LowKeyDevs retweeted
first look at the scribe studio blog-writing experience.
i overhauled the entire interface, then obsessed over the details that make writing feel stupidly good:
> buttery live preview
> your actual site typography
> instant desktop, tablet, and mobile views
> edits that follow you through long articles
> markdown stays the source of truth
no cms. no proprietary format. just your blog, finally pleasant to write.