@nightshiftdotsh is a young project with ambitious goals. We want to be the serverless runtime for AI agents and provide an incredible dev and operator experience. We're not yet, but with the help of the community we'll get there. Is and will always be open source. We're building software on github github.com/nightshiftco/nigh… and building community on slack shorturl.at/hphTc

I think it would be great if people were upfront about declaring their own understanding of a topic / their pull request. Now that everybody can talk confident with their clanker it becomes way too hard to understand if they knew what they were doing when they prompted it :(

Open > closed

$150k of angle funding to get us here nightshift.sh An agent harness that scales.

lots of discourse on context and not nearly enough on making search access better/more scalable for agents

Is there a serverless type platform where I can bring a LoRA adpater and inference is just handled for me?

why does docker desktop suck so bad

Our current runtime is held together with duct tape and bubblegum. By that of course I mean Python asyncio and sheer determination. We're starting the process of embedding containerd directly into Nightshift to standardize the communication between the Agent API Runtime (nightshiftd) and the sandbox. The goal is to have a flexible platform that can run agents in any sandbox you want (Kata Containers, Firecracker, runc, whatever) across any cloud or on-prem, inside or out of Kubernetes. I'm documenting this journey and as our first installment. I wrote up how I wrapped my head around container networking. nightshift.sh/blog/container…

In the latest release of @nightshiftdotsh (v0.4.3) Agent events are durable and repayable. Deploy your own KVM backed Agent Sandbox Platform. Is and will always be OSS.

The @nightshiftdotsh SDK is boring. We designed it that way. Write your agent in the frameworks you're used to (we're using the @OpenAI Agent SDK here). Wrap your agent in a decorator. Run it in a sandbox. Deploy and serve users. Is and will always be open source, for the community.

I'm new to async in python and I think I like this...will report back

We're running real business workloads through Nightshift 🦜. Define your task, connect your data, run jobs to completion. This is the next generation of compute, domain experts programming serverless work through (the hottest programming language) English! - Powerful online analytics processing - Agents run data driven business tasks to completion - Runtime Agent Evals guide them towards user intent - Git as the communication channel - Hardware level security We've launched the private beta of our managed service. If you're interested in chatting with us, we'd love to hear from you. lnkd.in/e3CnuDUZ We're open source! Join our community lnkd.in/eNhXXyYz

In the latest developments of @nightshiftdotsh we now ship Gollum (a simple wiki system built on top of git) that gives you a view into the agent workspace. This gives a consistent space for humans and agents to collaborate on knowledge work asynchronously. Have the agent read from documentation you add, or have them write out to a reports directory after performing a long running background job.

In the latest @nightshiftdotsh (v0.1.3) you can now turn a coding agent like @opencode into background knowledge worker. - Define your task spec on the job board - Run multiple tasks at once - Agents run tasks to completion - Run tasks in the background We're open source.

When GH is down it’s time to take a break 😌

so this ralph loop stuff, am I right?

so i guess I need to support windows....i don't wanna tho

We started @nightshiftdotsh to distribute coding agents (eg. Claude Code, @opencode, etc) to non-technical users as General Automation Machines. Yeah, we got business users to use the terminal... github.com/nightshiftco/nigh…