adobrawy

AI agents. Chatbot session of 8 hours is a lot. 16 vCPU might be useful when developing heavy application and agent need run application tests. You can think what infrastructure https://claude.ai/code needs.

I'm a Claude Code Web fan and a rather heavy user. So I was interested in your product. However, I couldn't find an answer on the website. What parts did you find so good that you ported them?

There was already attempt for dav1d to re-implement in rust by rav1d. It was hard to match performance: https://www.memorysafety.org/blog/rav1d-perf-bounty/

Is it what CloudFlare Organization is about? https://blog.cloudflare.com/organizations-beta/

If the problem is excessive deployments via GitHub Actions, why not use concurrency control on GitHub Actions ( https://docs.github.com/en/actions/how-tos/write-workflows/c... ) instead of relying on agent randomness and the hope that it won't make the same mistake again? Am I missing something?

Of course, this scheduled can be implemented on top of existing tools. However, it's incredibly convenient to have a UI where:

- you can easily edit the prompt

- you can see the prompt execution history

- you don't need any infrastructure to orchestrate it

- it works even when your computer is off

Once you start using it, it turns out to be very convenient.

That feature was silent launched about week ago for me.

I use it to:

- perform review of latest changes of code to update my documentation (security policies, user documentation etc.)

- perform review to latest changes of code, triage them, deduplicate and improve code - I review them, close them with comments for over-engoneering / add review for auto-fix

- perform review of open GitHub issue with label, select the one with highest impact, comment with rationale, implement it and make pull request - I wake up and I have a few pull request to fix issues that I can approve /finish in existing Claude Code thread

I want also use it to: - review recent Sentry issues, make GitHub issues for the one with highest priority, make pull request with proposed fix - I can just wake up and see that some crash is ready to be resolved

Limit of 3 scheduled jobs is pretty impactful, but playing with it give me a nice idea on how I can reduce my manual work.

CloudFlare for their serverless offering did it, and it works decent.

How are you performing backup of FerratDB? Are you using MongoDB tools, or are you using PostgreSQL-specific tools?

Nowadays universal answer for "what? why?" is AI. AI agent needs VMs to run generated code in sandbox as they can not be trusted.

You can install VSCode extension and use "/ide" to connect them.

Snowflake uses it as primary database for their metadata. https://www.snowflake.com/en/blog/how-foundationdb-powers-sn...

They rely on residential proxies powered by botnets — often built by compromising IoT devices (see: https://krebsonsecurity.com/2025/10/aisuru-botnet-shifts-fro... ). In other words, many AI startups — along with the corporations and VC funds backing them — are indirectly financing criminal botnets.

The best is to refer to official Tor project documentation for .onion over https: https://community.torproject.org/onion-services/advanced/htt...

tl;dr: Pressure from browsers, enterprise, and the overall ecosystem to use HTTPS (e.g., unavailability of advanced web features without HTTPS) is pushing for the use of HTTPS without exception, even for .onion sites with no significant technical advantage.

MCP also started as JSON-RPC over stdio. With solutions like GitHub Codespaces, devcontainers, or "background agents", I wonder if we'll see the development of JSON over SSE.

Currently, my environment uses Claude Code on bare metal, and my application runs in a container, and the agent can do "docker compose exec backend" without any restrictions (YOLO).

My biggest obstacles to adopting workflows with git worktree are the need to share the database engine (local resource constrains) and the initial migration time. Offloading to cloud might be interesting for that.

You can use Claude hooks ( https://docs.anthropic.com/en/docs/claude-code/hooks#stop-an... ) to trigger jj / git automatically.

In meantime, I recommend to read their availability reports on their blog, published monthly: https://github.blog/news-insights/company-news/

Yea, the reason is RFC9309.

In other words, are you using LLM as a text classifier?

At my company we use git squash PR merge strategy. This makes individual commits irrelevant, but PRs as a whole do. I use git town for stacked PRs. It's very nice to do another brunch when I've finished a logical stage, because small changes are reviewed and I merge often. When I have fixes, "git town sync" propagates up the stack automatically

It looks promising.

I retrieve error response when fetching exercise:

0:{"a":"$@1","f":"","b":"eETmgndxtv4Ar0i8Wync1"} 1:{"serverError":"An unexpected error occurred."}

My request: curl 'https://workout.cool/' \ -H 'accept: text/x-component' \ -H 'accept-language: en-US,en;q=0.9,pl-PL;q=0.8,pl;q=0.7' \ -H 'cache-control: no-cache' \ -H 'content-type: text/plain;charset=UTF-8' \ -b 'Next-Locale=en; _fbp=fb.1.1750253718188.954698194752805529' \ -H 'next-action: 7f80b017f78704b00d2411aebde5ba8318b475de6d' \ -H 'next-router-state-tree: %5B%22%22%2C%7B%22children%22%3A%5B%5B%22locale%22%2C%22en%22%2C%22d%22%5D%2C%7B%22children%22%3A%5B%22__PAGE__%22%2C%7B%7D%2C%22%2F%22%2C%22refresh%22%5D%7D%2Cnull%2Cnull%2Ctrue%5D%7D%5D' \ -H 'origin: https://workout.cool' \ -H 'pragma: no-cache' \ -H 'priority: u=1, i' \ -H 'referer: https://workout.cool/' \ -H 'sec-ch-ua: "Google Chrome";v="137", "Chromium";v="137", "Not/A)Brand";v="24"' \ -H 'sec-ch-ua-mobile: ?1' \ -H 'sec-ch-ua-platform: "Android"' \ -H 'sec-fetch-dest: empty' \ -H 'sec-fetch-mode: cors' \ -H 'sec-fetch-site: same-origin' \ -H 'user-agent: Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/137.0.0.0 Mobile Safari/537.36' \ --data-raw '[{"equipment":["PULLUP_BAR","BANDS","BODY_ONLY"],"muscles":["TRAPS","BACK","SHOULDERS","TRICEPS","FOREARMS","GLUTES","HAMSTRINGS","CALVES"],"limit":3}]'

You can use a self-hosted runner with an image that has anything pre-loaded.

I use an Apple device on a work device and I don't need to use Apple ID for anything. Apple laptop is the only device in the ecosystem.

Apple Store doesn't have useful apps for me, I rather use brew for installation, or I need to download dmg.

I don't use Apple Music because I use Spotify with my friends.

I don't use Apple Mail because we are in the Google ecosystem, so I use Gmail.

I don't use Apple Calendar (there is such a thing?), because in the Google ecosystem I use Google Calendar and - for notifications in the tray - Notion Calendar.

What do I lose by not using Apple ID?

I can see is:

404: NOT_FOUND

Code: DEPLOYMENT_NOT_FOUND

ID: arn1::llgmh-1743253533406-1fe7598253c0

I hope there is an author of the post. I appreciate sharing this example.

However, there is a caveat regarding the workflow code itself, that placing "run" statement with specific commands directly in YAML may make it difficult to debug locally. I would recommend that "run" statements be only for script execution and their number minimized. This way, scripts can be run locally without waiting for the entire CI pipeline to execute each time.

There is an open-source proxy for PostgreSQL with DocumentDB to implement the MongoDB interface: https://blog.ferretdb.io/ferretdb-v2-ga-open-source-mongodb-...

You do not need logical database isolation if you have even stronger isolation with virtual clusters. The pricing and quota will be important factor for that use case.

You can hide appointment schedules in desktop view: https://workspaceupdates.googleblog.com/2023/10/new-google-c...

I'm not sure what operations you're doing that you're CPU bound.

When exporting Gmail and hitting an execution timeout, I would implement partitioning by date. Google Apps Script has a lock mechanism ( https://developers.google.com/apps-script/reference/lock ), and you can use user properties for progress tracking ( https://developers.google.com/apps-script/reference/properti...() ). Therefore, you can schedule a task every x minutes to pull x days of data and keep the cursor in user properties.

@kwanbix, do you use hosted cal.com or self-hosted? If self-hosted, where is it hosted? How much time does it require for upkeep yearly? What is the infra cost?