DEV Community: Doug Trier

What AI tools do you use?

Doug Trier — Thu, 02 Jul 2026 01:08:49 +0000

I was watching YouTube videos to see what tools everyone else was using. There are a lot of good AI applications out there and I tried most of them.

When it came to IDEs I spent time with VS Code, Cursor, and Antigravity. Each one had things I liked but I kept running into the same problem. I wanted to use my choice of models.

I didnt want to create another account just to access a specific model.
I didnt want to juggle API keys or install plugins on every application.
I didnt want five different desktop apps from five different companies just to get basic workflows working.

So I built the application I wanted to use.

That project became More AI an application I’ve been working on

My goal was to build a desktop platform that is completely model‑agnostic and gives me every option I could ever want. Whether you want to use local models, API models, or models you already have installed through CLI.

I also wanted to make AI easier.

Instead of hunting down models, switching between apps, and figuring out which tool supports which provider, I wanted everything available from one place.

And More AI isnt just an IDE.

It includes Chat, an AI IDE, orchestration, voice, collaboration, image generation, music generation, video editing, document creator, and more — all designed to work together instead of living in separate applications.

I also built an AI Colosseum with different types of games to stress‑test models, especially local ones. There are great benchmark sites out there, but I wanted something interactive something you can play yourself, or run CPU vs model, model vs model, or even friend vs friend through peer‑to‑peer.

The peer to peer system I think can be useful. Imagine you and a friend coding together in a shared workspace then jumping into a game while your agents continue working in the IDE. Every surface is persistent. You can collaborate, test models, or just hang out and you can even stream it live if you want.

Security was also important to me.

So I built governance, audit trails, and execution history into the platform so I have visibility into what AI is doing at all times. and I have governance rules at the application level so agents cant rewrite them.

I originally built More AI because it’s the application I wanted for myself. Now I’m releasing it as open source maybe someone else will find it useful too.

I’d love to hear what AI tools you’re using today and what features you wish they had. If you want to check out what I’ve been working on or download it, you can get it on GitHub below.

It’s a big download, but it’s install‑ready. If you already have models, it will import them automatically. You can download more inside the app, or keep using your existing workflow.

DougTrier / MoreAI

Local-first Windows AI platform — 30+ workspaces in one app. Chat, IDE, Music, Voice, Games, Image, Video & more. Runs local, cloud optional. Free & open source (MIT).

More AI by Trier OS

The BYOB of AI: bring your own model, backend, cloud key, or local runtime. More AI keeps it governed, traceable, and under your control.

More AI is a free, open-source Windows desktop application for people who want one serious AI workspace without surrendering choice. Chat with cloud APIs, run local LLMs, code with tool agents, create music, images, video, documents, and workflows, orchestrate swarms, and audit what happened afterward. Every execution runs through a governance engine so nothing happens silently. No subscriptions. No API surprises. Run fully local, use cloud APIs, or mix both — you decide.

The goal is simple: make powerful AI feel usable, inspectable, and yours — no command-line setup required. Download the installer, run it, and the app handles everything else. The built-in terminal is there when you want it, but you should not need one just to get started.

Built…

View on GitHub

Can we have a little Phun?

Doug Trier — Sun, 10 May 2026 01:35:49 +0000

Was messing around with Suno and thought I would write a Phunny IT related song. Here is CRT reboot.

Fantasty Football feedback

Doug Trier — Tue, 28 Apr 2026 23:53:29 +0000

I would like to say first off; I have never played Fantasy football or been on a Fantasy football website. When I designed this application, I did it by reading what the rules were and back in the day...way back that is I use to play football. Without being adulterated by seeing what the designs were out there it was completely from my own concept. I have had a lot of downloads on it but no feedback.

 Honestly what I would like is feedback on how it looks, how close to reality it is. If you play fantasy football, is it useful as a scouting tool also? Any feedback would be appreciated.

if you are interested in looking at it and providing some feedback you can get it here on Github. It is open source and free

https://github.com/DougTrier/Trier_Fantasy_Football/releases/tag/v3.3.1

-Doug

Fantasy Football Anyone?

Doug Trier — Wed, 15 Apr 2026 17:20:01 +0000

Most of my time lately has been focused on building Trier OS — a deterministic, local-first platform for industrial operations.

But I’ve always believed the best systems aren’t built in one domain.

So in parallel, I’ve been working on something very different:

Trier Fantasy Football 🃏

It started as a side project, but it turned into a great way to experiment with:

interaction design (trading card UI)
local-first architecture
peer-to-peer synchronization (no central server)

Same principles, different environment.

And honestly — it’s been one of the most fun parts of the process.

Curious what people think, especially from a UI/UX and systems perspective.

https://github.com/DougTrier/Trier_Fantasy_Football/releases/tag/v1.0.0

New Build & Enterprise workflows - Trier OS

Doug Trier — Tue, 14 Apr 2026 04:09:19 +0000

Shipped another Trier OS build — looking for real feedback

At this point it’s less about “can it work” and more about “will it hold up in the real world.”

What has been tested in house:

Chaos testing across ~600 endpoints
Stability under bad conditions (offline, latency, partial failure)

Curious from others who’ve built large systems:

What actually broke when you scaled past “it works”?
What’s your approach to real stress testing (not just happy path)?

I have been fortunate enough to have also trained drivers on DSD applications across 3000 ends points so I understand stability and how important it is. I was the one that would break the application and report back to Dev's because I believe happy paths are not the real world.

With version 3.4.0

Scan a QR code at the machine → instantly pull up the work order → one tap to:

start work
complete
transfer
or put on hold (with reason codes)

No searching. No navigating menus.

The goal was simple:

reduce friction at the point of work to near zero

In a plant environment, even small delays add up fast — so this was designed to be as close to “scan and act” as possible.

Curious how others are handling this:

Are you tying assets directly to QR/NFC?
Are there and PWA issues that you have ran into and what were your solutions?

Repo:
https://github.com/DougTrier/trier-os

I built an industrial OS - here's what 33 years on the plant floor taught me about software

Doug Trier — Wed, 08 Apr 2026 00:19:27 +0000

The Problem Nobody Was Solving
I spent 33 years doing every job on the plant floor. Operator, supervisor, manager — I've done it all, hands-on.

And for 33 years I watched the same thing happen over and over: plants spending a fortune on enterprise software that the people actually running the facility couldn't stand using and would op out.

Until now.

What I Built
Trier OS is a full-stack, open-source plant operations platform. Not a SaaS. Not a subscription. A self-hosted, fully air-gapped system built specifically for heavy industry.

Here's what's under the hood:

Built-in Monaco IDE — authorized operators can write, sandbox, and hot-reload code directly inside the running production app
Deterministic Simulation Engine — replays historical plant event logs against your code changes before deploying. Mathematical proof it won't break the line.
Friction Cost Engine — calculates the exact financial cost of UI changes before they ship
3D GIS Mapping — Cesium-powered spatial intelligence to track assets across entire campuses
Mobile Barcode Scanning — WebRTC-based, works on iOS, Android, and Zebra rugged devices
EDR-Safe Local Mode — runs completely disconnected from the cloud on better-sqlite3
Full Rollback
Offline PWA
Full CMMS/MES Hybrid
349,000 lines of code. 1,378 modules. Everything documented.

How AI Made It Possible
With AI assistance I was able to build this in 33 days. But here's the thing — AI didn't know what to build. That came from three decades of watching plants operate, fail, and adapt.

The domain knowledge, the workflow logic, every edge case — that's 33 years of floors, not prompts. AI was the tool. The experience was the blueprint.

The Real Goal
Enterprise software licenses cost companies a fortune. By open sourcing Trier OS, my hope is simple: take those savings and hire in-house developers to run and customize it on site.

Free software. In-house talent. More jobs for programmers. Better tools for operators.

Win for the industry. Win for developers.

Happy to answer any questions
Get here https://github.com/DougTrier/trier-os