DEV Community

#dailybuild2026

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
I built an oracle that refuses to answer

I built an oracle that refuses to answer

Comments
7 min read
The CLI is the judge: compiling an org policy into a real `agents-cli` extension

The CLI is the judge: compiling an org policy into a real `agents-cli` extension

Comments
9 min read
Sloth Mode: an agent that does your dread while you do nothing

Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady Submission 🦥

Sloth Mode: an agent that does your dread while you do nothing

Comments
8 min read
The Enforcement Gap: measuring whether a coding agent follows its own rulebook

The Enforcement Gap: measuring whether a coding agent follows its own rulebook

1
Comments 2
9 min read
Ask in Telugu, get the answer out of an English insurance policy

Ask in Telugu, get the answer out of an English insurance policy

Comments
4 min read
A 2B vision model judged my listing photos. It knew what was wrong and still could not tell me whether to ship them.

A 2B vision model judged my listing photos. It knew what was wrong and still could not tell me whether to ship them.

1
Comments
5 min read
The model scored 19 percent, then rewrote its own manual to 100

The model scored 19 percent, then rewrote its own manual to 100

Comments
8 min read
How I made a small decision model's belief visible, and what that taught me about building honest AI demos.

How I made a small decision model's belief visible, and what that taught me about building honest AI demos.

Comments
7 min read
Four AI agents argue over one Penpot canvas until the design is good

Four AI agents argue over one Penpot canvas until the design is good

Comments
9 min read
Your agent says it addressed the review comments. Here's how you find out if it actually did.

Your agent says it addressed the review comments. Here's how you find out if it actually did.

1
Comments
5 min read
I built an app where you type a sentence and a real, editable design appears on a Penpot canvas

I built an app where you type a sentence and a real, editable design appears on a Penpot canvas

1
Comments
7 min read
Building a Rubik's Cube Decision Arena with 6 Decision Models (Laya, Clef, GLiNER 2.5, Kev and Strands Decider)

Building a Rubik's Cube Decision Arena with 6 Decision Models (Laya, Clef, GLiNER 2.5, Kev and Strands Decider)

Comments
14 min read
Building Anvil: code-as-action with a capability sandbox that explains its refusals.

Building Anvil: code-as-action with a capability sandbox that explains its refusals.

1
Comments
8 min read
How Arbiter works, why it was built, and the four measurements it made about an LLM judge that I had to look at twice.

How Arbiter works, why it was built, and the four measurements it made about an LLM judge that I had to look at twice.

Comments
12 min read
How I built a time-travel debugger for LLM agents on Burr, and what "replay the unchanged branch and prove nothing changed" actually takes.

How I built a time-travel debugger for LLM agents on Burr, and what "replay the unchanged branch and prove nothing changed" actually takes.

Comments
8 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.