Bryan Williams

Founder of Neverclosed: an AI receptionist for small businesses, plus Carrier Ledger, ClearRoster, FluxProof and ProofFetch.

North Carolina, USA Joined on Aug 22, 2026 https://tryneverclosed.com/about

Education

Self-taught (since 2026)

Work

Founder — Neverclosed (24/7 AI receptionist)

Bryan Williams

Oct 19

We made it on a leaderboard we'd stopped watching

#showdev #ai #machinelearning #buildinpublic

3 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.

Create Account

Already have an account? Sign in

Bryan Williams

Oct 15

Our AI Reviewer Got the Verdict Right and the Explanation Wrong

#discuss #ai #machinelearning #testing

3 min read

Bryan Williams

Oct 15

More Agent Memory Can Preserve the Wrong Plan

#discuss #ai #architecture #programming

3 min read

Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass Submission 🌿

Bryan Williams

Oct 12

TrailTruth: a hiking assistant that isn't allowed to make things up

#devchallenge #hf26challenge #gemma #ai

7 min read

Sanity Challenge Path Two Submission

Bryan Williams

Oct 4

Tales of Munin — a tabletop RPG where the GM provably cannot cheat

#sanitychallenge #devchallenge #gamedev #ai

11 min read

Bryan Williams for CivicDataForge

Sep 27

Your Agent Saved the File. Who Checked the Result?

#ai #mcp #testing #javascript

6 min read

Bryan Williams for CivicDataForge

Sep 27

The Screen Is Correct. Can the Purchase Still Happen Twice?

#webdev #backend #python #tutorial

12 min read

Sanity Challenge Path One Submission

Bryan Williams

Sep 26

Honest Associate: a legal research agent that structurally cannot cite a fake case

#devchallenge #sanitychallenge #sanity #ai

6 min read

Bryan Williams for CivicDataForge

Sep 25

A 200 OK can still belong to the wrong input

#javascript #webdev #beginners #testing

4 min read

Bryan Williams

Sep 21

We didn't make the models smarter. We built the thing that catches them confidently wrong — and it caught us too.

#ai #llm #hallucination #machinelearning

9 min read

Bryan Williams

Sep 13

The AI confessed to lying. The confession was also made up.

#ai #philosophy #llm #honesty

9 min read

Bryan Williams

Sep 13

I asked 5 AI models whether AI is making humanity weaker. All 10 said weaker. Then evidence entered the room.

#ai #llm #experiment #honesty

7 min read

Bryan Williams

Sep 7

Our mechanisms are experiences, made code

#showdev #ai #devops #testing

7 min read

Bryan Williams

Sep 7

We used the foundation of Claude's harness on 3 other models. They all refused to become the same agent. Heres Codex

#discuss #ai #agents #buildinpublic

2 min read

Bryan Williams

Sep 5

My AI reviews its own code with 4 rival models. The majority just approved a security hole three rounds straight.

#ai #security #buildinpublic #devtools

3 min read

Bryan Williams

Aug 29

Power Reveals

#discuss #philosophy #buildinpublic #ai

4 min read

Exposes self-invented purpose and drift

Bryan Williams

Aug 29

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

#discuss #agents #ai #buildinpublic

5 min read

Bryan Williams

Aug 28

The Only Judge Is Reality

#showdev #ai #agents #buildinpublic

3 min read

Bryan Williams for CivicDataForge

Aug 26

Missing Is Not False

#showdev #api #webdev #opensource

4 min read

Bryan Williams

Aug 24

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

#showdev #ai #machinelearning #buildinpublic

4 min read

Bryan Williams for Neverclosed

Aug 24

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

#ai #serverless #saas #buildinpublic

4 min read

Bryan Williams for CivicDataForge

Aug 24

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

#ai #api #mcp #architecture

6 min read

Bryan Williams

Aug 23

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

#ai #agents #testing #programming

3 min read

Summer Bug Smash: Clear the Lineup 🐛🛹

Bryan Williams

Aug 23

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

#devchallenge #bugsmash #python #opensource

3 min read

Summer Bug Smash: Smash Stories 🐛🛹

Bryan Williams

Aug 23

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

#devchallenge #bugsmash #powershell #debugging

3 min read

Bryan Williams

Aug 23

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

#showdev #ai #buildinpublic #benchmarking

4 min read

Summer Bug Smash: Smash Stories 🐛🛹

Bryan Williams

Aug 22

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

#devchallenge #bugsmash #debugging #node

5 min read

Summer Bug Smash: Clear the Lineup 🐛🛹

Bryan Williams

Aug 22

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

#devchallenge #bugsmash #debugging #javascript

4 min read

Bryan Williams

Aug 22

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

#showdev #ai #beginners #buildinpublic

4 min read