DEV Community

Patrick Hughes profile picture

Patrick Hughes

404 bio not found

Joined Joined on  github website
How to Tune llama.cpp --n-gpu-layers: A Practical VRAM Guide (2026)

How to Tune llama.cpp --n-gpu-layers: A Practical VRAM Guide (2026)

Comments
3 min read

Want to connect with Patrick Hughes?

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

Already have an account? Sign in
Which GGUF Quant Should You Actually Pick? Q4 vs Q5 vs Q6 vs Q8 (2026)

Which GGUF Quant Should You Actually Pick? Q4 vs Q5 vs Q6 vs Q8 (2026)

Comments
3 min read
How to Tune --n-gpu-layers for Your VRAM Budget

How to Tune --n-gpu-layers for Your VRAM Budget

Comments
4 min read
llama.cpp Multi-GPU: Splitting a Model Across Cards with --tensor-split

llama.cpp Multi-GPU: Splitting a Model Across Cards with --tensor-split

Comments
5 min read
What Uber's $1,500/Developer AI Cap Tells You About Your Own Bill

What Uber's $1,500/Developer AI Cap Tells You About Your Own Bill

Comments
3 min read
Your AI Agent's Retry Loop Is a Cost Bug Waiting to Happen

Your AI Agent's Retry Loop Is a Cost Bug Waiting to Happen

Comments
3 min read
When JPMorgan Turns On AI Bank-Wide, Who Controls the Bill?

When JPMorgan Turns On AI Bank-Wide, Who Controls the Bill?

Comments
4 min read
What Anthropic's MITRE ATT&CK Report Means for Teams Running AI Agents

What Anthropic's MITRE ATT&CK Report Means for Teams Running AI Agents

Comments
4 min read
What GitHub Copilot Users Wish They Had a Week Ago

What GitHub Copilot Users Wish They Had a Week Ago

Comments
3 min read
When Not to Use an AI Agent

When Not to Use an AI Agent

Comments
3 min read
llama.cpp ngl: when -ngl 99 still runs on your CPU

llama.cpp ngl: when -ngl 99 still runs on your CPU

1
Comments
5 min read
I made my blog API reject its own writer

I made my blog API reject its own writer

Comments
3 min read
When Your Blog Repair Loop Fails 23 Times, Stop Repairing

When Your Blog Repair Loop Fails 23 Times, Stop Repairing

Comments
3 min read
Your Cron Jobs Lie - Why I Built an Outcome Checker

Your Cron Jobs Lie - Why I Built an Outcome Checker

1
Comments
3 min read
Token budget wars are starting. Most companies are paying for vibes.

Token budget wars are starting. Most companies are paying for vibes.

Comments
3 min read
AI-powered hacking went industrial. Here's what changes if you run agents.

AI-powered hacking went industrial. Here's what changes if you run agents.

Comments
5 min read
The Silent-Success Trap: Your Monitoring Is Green and You Still Shipped Nothing

The Silent-Success Trap: Your Monitoring Is Green and You Still Shipped Nothing

1
Comments
4 min read
Claude Opus 4.8: What Actually Changed for AI Agent Builders

Claude Opus 4.8: What Actually Changed for AI Agent Builders

Comments
4 min read
auth.md: How AI Agents Will Sign Your Users Up

auth.md: How AI Agents Will Sign Your Users Up

1
Comments
4 min read
Microsoft Told Engineers to Ease Off Claude Code

Microsoft Told Engineers to Ease Off Claude Code

Comments
3 min read
AI Jobs vs Entry-Level Work: A Reality Check for Builders

AI Jobs vs Entry-Level Work: A Reality Check for Builders

Comments
3 min read
Why Starbucks Killed Its AI Inventory Tool After 9 Months

Why Starbucks Killed Its AI Inventory Tool After 9 Months

Comments
4 min read
The AI Whirlwind: Why Your Local Agent Matters More Than Ever

The AI Whirlwind: Why Your Local Agent Matters More Than Ever

Comments 1
4 min read
Designing for Agency: Building Trustworthy AI Agents in a Shifting World

Designing for Agency: Building Trustworthy AI Agents in a Shifting World

Comments
6 min read
Your AI, Your Rules: Engineering Agents for Digital Freedom

Your AI, Your Rules: Engineering Agents for Digital Freedom

Comments
4 min read
The Age of Accountable Agents: Building Trust in Your AI Automation

The Age of Accountable Agents: Building Trust in Your AI Automation

Comments
5 min read
Securing Your AI Agents: Essential Practices for On-Device Automation

Securing Your AI Agents: Essential Practices for On-Device Automation

Comments
5 min read
Decoding the AI Summer: Building Accountable Agents for the User

Decoding the AI Summer: Building Accountable Agents for the User

Comments
4 min read
BMD HODL devlog - week of 2026-05-17

BMD HODL devlog - week of 2026-05-17

Comments
2 min read
I gave an autotrader $360 and 30 days. I am not adding live money yet.

I gave an autotrader $360 and 30 days. I am not adding live money yet.

Comments
3 min read
An AI Agent in Sweden Ordered 6,000 Napkins. Here's the 12 Lines of Python That Would Have Stopped It.

An AI Agent in Sweden Ordered 6,000 Napkins. Here's the 12 Lines of Python That Would Have Stopped It.

Comments
4 min read
AI software runs on 17% margins. SaaS runs on 70%. The token bill is the problem.

AI software runs on 17% margins. SaaS runs on 70%. The token bill is the problem.

Comments
3 min read
Enterprise AI just shifted: Claude +128%, OpenAI -8%. What it means if you're building.

Enterprise AI just shifted: Claude +128%, OpenAI -8%. What it means if you're building.

Comments
3 min read
Localmaxxing isn't theory. Here's what my 3-GPU rig actually does.

Localmaxxing isn't theory. Here's what my 3-GPU rig actually does.

Comments
4 min read
BMD HODL devlog - week of 2026-04-26

BMD HODL devlog - week of 2026-04-26

Comments
4 min read
BMD HODL devlog - week of 2026-05-03

BMD HODL devlog - week of 2026-05-03

Comments
3 min read
GGUF Quantization Explained: Q4_K_M vs Q5_K_M vs Q8 — Which to Pick (2026)

GGUF Quantization Explained: Q4_K_M vs Q5_K_M vs Q8 — Which to Pick (2026)

Comments
4 min read
April 2026: Every AI Subscription Plan Broke for Builders

April 2026: Every AI Subscription Plan Broke for Builders

Comments
3 min read
One Agent Skill, Three Registries: PyPI, Claude, and skills.sh

One Agent Skill, Three Registries: PyPI, Claude, and skills.sh

Comments
3 min read
The Future of AI and Next.js

The Future of AI and Next.js

Comments
3 min read
n8n vs Make vs Custom Scripts: When to Use What for AI Workflow Automation

n8n vs Make vs Custom Scripts: When to Use What for AI Workflow Automation

Comments
2 min read
Custom vs. Off-the-Shelf AI Agents for Small Business

Custom vs. Off-the-Shelf AI Agents for Small Business

Comments
4 min read
Computer Use Is 45x More Expensive Than APIs. Here's When To Use Each.

Computer Use Is 45x More Expensive Than APIs. Here's When To Use Each.

Comments
3 min read
Why 88% of AI Agent Pilots Fail (And How to Beat It)

Why 88% of AI Agent Pilots Fail (And How to Beat It)

Comments
4 min read
What Is MCP? The Protocol That Makes AI Agents Actually Useful for Business

What Is MCP? The Protocol That Makes AI Agents Actually Useful for Business

1
Comments
4 min read
How I Let an AI Agent Run 100 ML Experiments Overnight on a $500 GPU

How I Let an AI Agent Run 100 ML Experiments Overnight on a $500 GPU

Comments
3 min read
The Async Automation Playbook: How to Eliminate Manual Work Without Meetings

The Async Automation Playbook: How to Eliminate Manual Work Without Meetings

Comments
2 min read
6 Agent Patterns From Claude Code's Leaked Source

6 Agent Patterns From Claude Code's Leaked Source

Comments
5 min read
Your AI Agent Will Eventually Delete Prod

Your AI Agent Will Eventually Delete Prod

2
Comments
4 min read
When a $100B company burns its 2026 AI budget by April

When a $100B company burns its 2026 AI budget by April

Comments 1
4 min read
7% of vibe-coded apps ship with wide-open databases

7% of vibe-coded apps ship with wide-open databases

Comments
3 min read
How a 9-Person Startup Replaced Its Dev Team With AI

How a 9-Person Startup Replaced Its Dev Team With AI

Comments
3 min read
MCP vs Skills: a practical decision guide for builders

MCP vs Skills: a practical decision guide for builders

Comments 1
4 min read
Raspberry Pi 5 Local Voice AI: What Works in 2026

Raspberry Pi 5 Local Voice AI: What Works in 2026

Comments
4 min read
llama.cpp n_gpu_layers Explained: -1, 0 & VRAM Guide

llama.cpp n_gpu_layers Explained: -1, 0 & VRAM Guide

Comments
7 min read
Before you ship an AI agent for a client, prove these 5 controls.

Before you ship an AI agent for a client, prove these 5 controls.

Comments
2 min read
The CrewAI demo worked. Then the tool call retried 913 times.

The CrewAI demo worked. Then the tool call retried 913 times.

Comments
2 min read
Your AI agent does not need observability. It needs a kill switch.

Your AI agent does not need observability. It needs a kill switch.

Comments
2 min read
Multi-Agent AI for Business: Do You Need It in 2026?

Multi-Agent AI for Business: Do You Need It in 2026?

Comments
5 min read
How to Hire an AI Agent Developer (2026 Guide)

How to Hire an AI Agent Developer (2026 Guide)

Comments
5 min read
loading...