Me too!

Image recognition

Edge gallery app

Edge gallery app downloading the model on device

Audio transcription and translation to Malayalam

Gemma-4 E4B running on my Google Pixel 8 phone. Impressive audio transcription, translation, image recognition. Knowledge retrieval not so much (understandably as small model).💯
The model is a lot more capable than gemma-3. It's time to update the post I wrote last year! annjose.com/post/mobile-...

Read HN

Start small, keep it simple. So you can get the benefit of the tool without feeling overwhelmed. As you are using it, you will see patterns, and only then look for plugins, extensions, advanced features.

The only tool that maybe useful from the beginning is Obsidian Web Clipper browser extension.

Digests

News

Public agencies

All notification groups (8)

Nextdoor notification settings:
- 8 topic groups
- each group has 6-7 topics (one has 26!)
- each topic has 3 toggles - in-app, email and push

That is 167 buttons to click! 😮
You can't turn off a group as a whole.

And the gift keeps on giving - every time I turn them off, new topics pop up 🤦‍♀️.

100% agree.

> Does it matter if it works even if it's hot garbage? Yes. Emphatically yes
> 37000 lines of code a day isn't a flex, it's a warning

Amen!

AI should enable us to do things we couldn't do before, by doing them better, not worse. Let it do the boring stuff that don't teach us anything.

The photo I have been waiting for - to dedicate to the small group of people in our society who are confused about the shape of our beautiful planet.

Claude being honest about what it doesn't know. 👏 Thank you!

> On the SHA pinning question — yes, I'm aware of the general trend of npm supply chain attacks. I don't have details on a specific recent axios incident since my knowledge cuts off May 2025, so I'd be making things up if I described it.

Yes, they need to be Atmospherified, not just verified. #atproto

LinkedIn download data page that shows the options to download your data. There are two options here: 1. Download larger data archive, including connections, verifications, contacts, account history, and information we infer about you based on your profile and activity. Learn more 2. Want something in particular? Select the data files youre most interested in. Articles Invitations Profile Recommendations Registration And a button 'Request archive'

TIL You can download your LinkedIn data as CSV from www.linkedin.com/mypreference...

I want to write a script to parse the csv, filter the connections in my network who I don't know (may have connected long ago). Then write a browser ext or Playwright script to unfollow/disconnect them on LinkedIn.

I would love to try it, but is in waitlist 😥. Hope I will be accepted soon!
It's a great idea - I see it as _my_ sky!

If you like the bottom nav bar with the back button, you can enable it System -> Navigation Mode, select 3-button navigation.
I switched from Apple to Pixel a few years and there is no going back. I feel Android is more functional than iOS (and I was a iOS developer).

Very true! I see this first-hand in coding.

Great post! I agree and see it first hand in coding. Thanks for sharing your workflow. Two gems:

> I don’t expect the robot to replace reporting, I expect it to take the pain out of reporting.

> If you’re a writer, Claude can make you a better writer. If you’re not a writer, it can’t do magic.

Basically, the inner content becomes literal text, not fence boundaries.

See more at spec.commonmark.org/0.31.2/#fenc...

Terminal showing two nested syntax blocks - the outer one with 4 ticks and inner one with 3 ticks to show the output text

TIL you can nest markdown code blocks by using more backticks in the outer fence than inner one.

I often want to document shell commands with their output - commands in ```sh, but output as plain text. I used to prefix each line with #. But now I use 4 backticks for outer block and 3 for the inner.

I think it should be both - bigger models for big/complex tasks vs smaller models for smaller/simpler tasks. In fact yesterday I tried sub agents and found a feasible workflow to do that. We can use /agents or /agent-teams - each has its own pros and cons.
code.claude.com/docs/en/agent-teams

TIL: Create custom agents in Claude Code using /agents command to handle smaller tasks with specific tools, and can even run on smaller models. 🔥

Invoke in session:
- @agent
<name> - guaranteed to run
- natural language - Claude decides

Or at launch shell command
- claude --agent <name>

what is Moob? tell me more.

The Agent Toolkit — OpenClaw Hack Day w/ OpenAI Codex · Luma Build the OpenClaw skills that should exist but don't. OpenClaw hit 160K+ GitHub stars by becoming the go-to framework for local AI agents. Its skill system…

Had a lot of fun at the OpenClaw hackathon at SF yesterday with @georgeck.me. Met a good mix of people - more senior devs than usual. We loved that the focus was to build something real. It nudged us to install OpenClaw on our laptop in a fully locked down docker container.

luma.com/openclaw-hac...

Tetris game running in browser - significant difference in the UI

Result from two harness/model side-by-side:
Prompt: write a tetris game in typescript with beautiful UI

Left: Claude Code + Sonnet 4.5
Right: Open Code + GLM-5

Harness matters. Model matters.
What would it be with Claude Code + GLM-5?

There is something beautiful about sitting with Claude Code and working on a problem together. ❤️

Every time I talk to it, I learn something new. No judgment, we just brainstorm, try things, test and deploy! #joy-of-building

The top toolbar in the Obsidian Clipper has a 'Reader' button (between the ... and Highlight buttons), which will triggered this reader.

@annjose - areyougoingexponential.rhys.dev @annjose has 2,453 contributions on GitHub. Ann Catherine Jose's GitHub contributions are down 9% over the last 6 months.

The line goes up and up 📈

2023: left corporate, went self-employed ❤️
2025: started building personal projects 🚀
2026: discovered agentic coding 🔥

Three life pivots, three peaks.

areyougoingexponential.rhys.dev/annjose

The 8 Levels of Agentic Engineering — Bassim Eledath AI's coding ability is outpacing our ability to wield it effectively. That gap closes in levels — 8 of them. Here's the progression from tab complete to autonomous agent teams.

Levels of Agentic Engineering
View Article | Join the HN Conversation

Summary of HN discussion 🧵👇

A screenshot of the VS Code "Timeline" view, which displays a chronological history of local file changes including Git commits, manual edits, undos, redos, and file saves.

Screenshot of Visual Studio Code showing the "NPM SCRIPTS" sidebar view, which is highlighted in yellow to show how it organizes scripts from all sub-packages into a single, consolidated list. The view displays folders for different packages (like "email-worker" and "extension") with their respective scripts (such as "build", "dev", and "test") listed underneath. A blue box highlights the "Run" and "Debug" icons next to a specific script, illustrating how to execute them directly from the sidebar.

Screenshot of Visual Studio Code showing how to enable hidden sidebar views. A blue arrow points to the top of the sidebar near "EXPLORER" with the instruction "right-click here." This action displays a context menu, labeled "you see this menu," where "Timeline" and "NPM Scripts" are listed. Checkmarks next to these options indicate they are currently enabled, illustrating how to toggle these views on or off.

TIL💎: VS Code has two super useful views in the left sidebar:

1) NPM SCRIPTS: shows all scripts across all sub-packages as one list. You can run / debug any script right there!🔥
2) TIMELINE: shows local history (undo, redo, saves) & git commits.💯

Right-click the EXPLORER header to turn them ON.

The @figma.com MCP server now supports a full roundtrip workflow from VS Code! Bring Figma designs into code, then capture your rendered UI back to the Figma canvas as editable frames.

Join us for a special livestream tomorrow at 9:30 am PT to see it live in action: youtube.com/live/O1i-Te7...

Published a new skill: prompt-review

It audits any prompt against Claude's official best practices - clarity, context, examples, structure. Outputs a scored review and a revised prompt. 🔥

It's part of my collection of agentic coding tools. Open source. Ready to use now!

github.com/annjose/agen...

We often hear "AI is like a junior / intern" as if they're just eager but dumb. It always felt off coz juniors are usually smarter than we credit.

I like this framing in Claude docs:
> Think of Claude as a brilliant but new employee who lacks context on your norms and workflows

The model context 💯

Played around with the new Nano Banana 2 and updated my avatar with an image it created.

I selected a pencil sketch theme, uploaded my color profile photo, and it generated this version. Pretty impressive.

blog.google/innovation-a...