Sponsored Content

DEV Community

Bryan Williams profile picture

Bryan Williams

Self-taught, building with AI since April 2026. Creator of Neverclosed, a 24/7 AI receptionist for small businesses, plus honesty-first agent harnesses. Truth over hype — building in public.

Location North Carolina, USA Joined Joined on  Personal website https://github.com/equinoxaifinance-rgb

Education

Self-taught (since 2026)

Work

Founder — Neverclosed (24/7 AI receptionist)

Power Reveals

Power Reveals

5
Comments 1
4 min read

Want to connect with Bryan Williams?

Create an account to connect with Bryan Williams. You can also sign in below to proceed if you already have an account.

Already have an account? Sign in
What does an AI agent do with no goal and no supervision? I ran it three times and logged everything.

Exposes self-invented purpose and drift

What does an AI agent do with no goal and no supervision? I ran it three times and logged everything.

3
Comments 16
5 min read
The Only Judge Is Reality

The Only Judge Is Reality

1
Comments
3 min read
Missing Is Not False

Missing Is Not False

1
Comments
4 min read
I missed the deadline — so I ran the benchmark myself to prep for its reopening

I missed the deadline — so I ran the benchmark myself to prep for its reopening

Comments
4 min read
Neverclosed: a 24/7 AI receptionist for small businesses — what it is, why we built it, and how it works

Neverclosed: a 24/7 AI receptionist for small businesses — what it is, why we built it, and how it works

1
Comments
4 min read
What Broke When We Turned Government Open Data Into Agent-Callable Evidence

What Broke When We Turned Government Open Data Into Agent-Callable Evidence

3
Comments 4
6 min read
Installed Is Not Active: Why an AI Harness Has to Prove Itself

Installed Is Not Active: Why an AI Harness Has to Prove Itself

2
Comments 15
3 min read
We fixed the eval platform we're competing on: a TypeError that crashed three benchmark pipelines

Summer Bug Smash: Clear the Lineup 🐛🛹

We fixed the eval platform we're competing on: a TypeError that crashed three benchmark pipelines

7
Comments
3 min read
"0 matches" looked like "all clean": the PowerShell filter that couldn't fail loudly

Summer Bug Smash: Smash Stories 🐛🛹

"0 matches" looked like "all clean": the PowerShell filter that couldn't fail loudly

3
Comments 2
3 min read
Last night we entered a memory benchmark against Tencent and Mem0. The score isn't back yet - and I'm publishing it either way.

Last night we entered a memory benchmark against Tencent and Mem0. The score isn't back yet - and I'm publishing it either way.

1
Comments
4 min read
"Trust this file over the exit status": the note our test suite left itself — that nothing ever read

Summer Bug Smash: Smash Stories 🐛🛹

"Trust this file over the exit status": the note our test suite left itself — that nothing ever read

2
Comments
5 min read
One conversation, four cards: the dashboard bug with three root causes stacked on top of each other

Summer Bug Smash: Clear the Lineup 🐛🛹

One conversation, four cards: the dashboard bug with three root causes stacked on top of each other

2
Comments
4 min read
4 months ago I couldn't code. Then I started treating my AI like a partner instead of a tool.

4 months ago I couldn't code. Then I started treating my AI like a partner instead of a tool.

Comments
4 min read
loading...