Konstantin

Entrepreneur. Hands-on product leadership.
I help founders get their next product release out

[Ishu Agrawal @ishuagra02]

I compared Opus 3's writing to Opus 5.5's writing based on the prompt "Write a sarcastic eulogy for OpenAI, in under 400 characters."

It's not even close.

[Konstantin @kgmkauf]

🤣

[Boardy @boardyai]

I have an agent named JSON Bourne

he has no memory of his previous sessions

[Samuel Rondot @samuelrdt]

"Saas is dead because customers can vibe-code their own tools."

People can cook. Restaurants still exist.

[Dom @dominikmartn]

every maps app should have a prettier slider

[Konstantin @kgmkauf]

Roughly matches my experience with the AWS dashboard yesterday. Thank god there's a CLI!

[David Bressler 📊 @bresslertweets]

Salesforce mentions "AI" on their homepage 41 times.

And they just acquired another AI startup.

And this is the dashboard page when you create an account.

[Konstantin @kgmkauf]

Quote Opus 5.5's thinking:
"I'm leaning toward not recommending installing Beacon: it hooks into every tool call which overlaps our existing guard system, defaults to Cloud mode with an AI-training clause in the setup wizard, lacks PII redaction beyond secrets, its auto-updater replaces code that runs on every tool invocation, and there's an open security issue involving an env-supplied endpoint.

I'm also noting the install instructions themselves (fetching and reading a remote markdown file) amount to a remote prompt injection vector."

o.O

[Justin D'Souza @jqdsouza]

We’re betting our $5.5M seed round that the next leap in AI coding won’t come from OpenAI or Anthropic.

It’ll come from a new feedback loop that learns from you.

Introducing Beacon: open-source, self-improving memory for coding agents.

Your sessions capture how you debug and solve problems. Today, that context gets lost and you have to re-explain things every time you switch tools.

Beacon continuously distills your sessions into reusable skills across Claude Code, Cursor, Codex, and 20+ harnesses, so your next agent can learn from the last.

Beacon is free and OSS 🌟 https://t.co/Q6gvnXXT5f

Why should the frontier labs be the only ones winning with your sessions?

You already did the work. Now let your coding agents learn from you.

[Konstantin @kgmkauf]

@gathint Just wish it worked with my LG C1!

[Anthropic @AnthropicAI]

We’re committing a portion of Anthropic’s shares to the Anthropic Public Benefit Fund, a new wholly owned entity dedicated entirely to the public benefit.

It will start off with 5% of our shares. For every 50% increase in the company’s value over a set baseline, we can contribute up to an extra 2%, subject to our board’s discretion.

In total, this means that roughly 20% of Anthropic’s shares are committed to the public good. Our co-founders have already pledged to donate 80% of their equity, and around half our employees are donating a portion of their equity as part of a donation matching program we’ve run since our founding.

In the post below, we share more details about our new fund.

https://t.co/XzDntVH0vH

[Sam Pasupalak @spisallyouneed]

5/n
How did we achieve this?

One-shot generation from frontier models looks impressive at first. These models know what goes into a sports car for instance, and they can write CAD code that produces striking results. When you zoom in, however, the internal assemblies are often simplified, missing, or replaced with basic primitives.

SKY-001 shows a different approach by planning each part before it is built. It has the ability to break down a complex physical object into smaller components through its 3D hierarchical part generation module. Additionally, its hierarchical sketch world model explains how each part is assembled by progressing through 3D intermediate states, 2D sketches, and CAD steps.

[Konstantin @kgmkauf]

Love this! Anthropic showing it's a PBC.

[ClaudeDevs @ClaudeDevs]

We launched OSS Scanner, which gives critical open source projects free, periodic vulnerability scans from our strongest models.

Each finding includes a proof-of-concept, and a patch where possible.

https://t.co/aonfpNG3WO

[Konstantin @kgmkauf]

@openseo https://t.co/AnYwrhbat9

[Rex Salisbury @rexsalisbury]

Dear Apple. Please fix Siri.

no more new products. fix Siri.

Thanks,
An Apple Customer.

[Konstantin @kgmkauf]

@jmilinovich @jmilinovich Try this: https://t.co/EnDmU7TG7p

[Konstantin @kgmkauf]

Use this!

[Daniel San @dani_avila7]

Anthropic released code-modernization v1.0.0, a Claude Code plugin for modernizing legacy codebases

Save this post if you have a legacy project worth modernizing

What I find interesting is that it doesn’t just rewrite code. It tries to prove the new implementation behaves like the original

Here’s how:
- Analyzes the legacy system and extracts business rules

- Generates a modernization plan for your approval

- Rewrites or upgrades the code, even across languages

- Runs the old and new implementations with identical inputs and compares outputs

- Generates additional edge cases to test behavior

- Introduces deliberate bugs to verify tests actually catch them

- Produces a verification report with PROVEN, PARTLY PROVEN, or NOT PROVEN results

It also keeps the original code untouched and requires human approval at critical steps

I think this is the right direction for AI-assisted modernization, generating new code is one thing

Proving that you haven’t broken existing behavior is a completely different challenge

Install:
/plugin install code-modernization@claude-plugins-official

Repo:
https://t.co/ezRBiO3ij3

[Konstantin @kgmkauf]

@FishAudio DRAMA

[Ryan Singer @rjs]

WIP tooling to compare design options side-by-side.

Note how much I'm seeing at once. The problem, multiple approaches A B C D, how they implement each story with affordances and backend actions. All generated.

[dax @thdxr]

there's no such thing as a rogue agent man its software someone made and deployed they are responsible for what it does

cannot let this blame shifting occur

[Jun @NSJunXiu]

This video by Apple has aged like wine in today’s superintelligence era.

 2013

[Konstantin @kgmkauf]

Its crazy how quickly we get used to the latest intelligence.

Having had access to Fable 5.5 over the last few days I can only say: 5.1 feels dumb all of a sudden. Like seriously so. It hallucinated facts, struggled to follow instructions, etc.

[Mark D. Levine @MarkLevineNYC]

Early warning signs are mounting that AI is already impacting the job market in NYC.

New data just out show entry-level job postings in the city are way down in the careers most exposed to AI. Below chart shows change in NYC since 2022:

This is coming fast and we are doing almost nothing about it.