Adrian Withy

Build meaningful things. Retweets are not endorsements. Tweets are my own.

[Merriam-Webster @MerriamWebster]

Why is it ‘cancelled’ in the U.K. but ‘canceled’ in the U.S.?

Because we gave them that L in 1776.

[dax @thdxr]

as good as LLMs have become at understanding instructions and solving the task, they still don't produce the code i would

their solutions are always the 2nd or 3rd best way of doing something

you think this doesn't matter because so are yours

[ThePrimeagen @ThePrimeagen]

never forget

The people who tell you coding is a solved problem also created MCP

[Adrian Withy @adrianwithy]

Fantastic talk https://t.co/KawU6iYGbS

[Mario Zechner @badlogicgames]

recommended reading. deepmind's new AI control roadmap. looks like they've given up on solving the lethal trifecta directly. the new direction seems to be a tower of LLMs. i suppose that's as good as it gets.

https://t.co/RiXTjnCJuz

[Armin Ronacher ⇌ @mitsuhiko]

Some more thoughts on looping in coding agents. https://t.co/loRLPZduec

[fks @FredKSchott]

Just Shipped: Flue 1.0 Beta

Flue is the TypeScript framework for building the next generation of agents, designed around an open agent harness with zero LLM lock-in. It’s like Astro, for agents.

Flue 1.0 has been redesigned around three core primitives:

🔁 Workflows — structured automations designed for background work, where your code drives the agent from start to finish.

🧭 Agents (New!) — autonomous, stateful loops where the model drives itself to complete a given task.

📡 Channels (New!) — connect agents to Slack, GitHub, Linear, Discord, Teams, and more. Flue handles the boilerplate for you.

Everything shares the same durable foundation, powered internally by Pi, Vite, and Durable Streams. Deploy anywhere, use any LLM, and recover running agents across restarts and downtime.

We’ve talked to a lot of teams building agents, and keep hearing the same thing: getting to production is hard work. We built Flue to help change that.

Flue 1.0 Beta is available today. Give it a try and let me know what you think!

[Gergely Orosz @GergelyOrosz]

So apparently after Meta leadership:

- Force reassigned some of the best devs on teams to AI data labelling fulltime

- Laid off another 10%

- Started to record every dev’s screen in the US 24/7

They now realized that it has, indeed started to destroy their eng culture. And are now trying to walk back.

All of the above was unprompted, not forced by anything external or even business reasons (Meta recorded record revenue, record profits)

The biggest self-inflicted eng culture destruction I’ve seen in a matter of weeks

[Romi Bean @Romi_Bean]

This Frenchman describing Buc-ee’s is the only thing you need today

(IG: Erosbrousson)

[Ethan Mollick @emollick]

Final one-shot prompt I did before the Fable interruption: "build me a cool simulation thing that lets me demo the various forms of FTL travel from both famous works of fiction and scientific speculation. it should be graphically compelling & interesting." https://t.co/j9QwssK1mD

[Evergreen Intel @vcdgf555]

It is an incredibly small world... Never taken pics or video of a plane takeoff and have it not return safely. Not a great feeling.

Aviation photography has been a blessing of a hobby, but as much as it can be a joy or exciting, these are, in the end, dangerous jobs people do every day. And I don't ever lose sight of that.

I am quite relieved to hear the pilot is ok. I wish them the best in recovery after ejecting: no small experience to go through.

[DanT @uyintans]

@GeorgeMayer Preventing models from finding and fixing bugs would break agentic coding

[Adrian Withy @adrianwithy]

@ClementDelangue I wouldnt be surprised if we see licensing requirements similar to amateur radio. There's no real way to fix it.

[Everlier @Everlier]

@LLMJunky @DavidSacks As they say in agentic security, there's no firewall for English.

[Nathan McNulty @NathanMcNulty]

@UK_Daniel_Card @ZackKorman How many people can diff a patch or read a CVE and create an exploit?

How many people with an LLM can do it?

Also, how quickly?

We're seeing more vulns on edge devices on a faster cadence, and I just think it's going to get worse before it gets better ;)

[Nathan McNulty @NathanMcNulty]

@adrianwithy @UK_Daniel_Card @ZackKorman Just because they can doesn't mean they do ;)

[Adrian Withy @adrianwithy]

@NathanMcNulty @UK_Daniel_Card @ZackKorman Edge device vendors also have LLMs and can white box test with them. Finding a vuln before the vendor does should be very expensive.

[Adrian Withy @adrianwithy]

@ZackKorman Not as bad as the hype would have one believe. Perhaps it might speed up the next couple of months and cause a short term spike in chaos, but I've seen nothing that says this is a fundamentally different dynamic than what we've experienced since December.

[Adrian Withy @adrianwithy]

My vibe coding journey: https://t.co/f6fXygcdYP

[Mario Zechner @badlogicgames]

watch me how i work with coding agents like a caveman.

https://t.co/vrY8zZAkeY