DEV Community

Mason Reed profile picture

Mason Reed

Software engineer focused on AI. Early walks with the dog, late attempts at woodworking. Happiest when something I built actually works.

Joined Joined on 
Changing Gemini CLI Directories: Workspace, Project Settings, and Global Config

Changing Gemini CLI Directories: Workspace, Project Settings, and Global Config

Comments
6 min read
Building an AI Companion: Persona, Memory, Tools, and the Tests That Matter

Building an AI Companion: Persona, Memory, Tools, and the Tests That Matter

Comments
6 min read
Top 6 OpenClaw Skills you can't afford to miss in 2026

Top 6 OpenClaw Skills you can't afford to miss in 2026

Comments
7 min read
How to Add AI Video Generation to a SaaS App

How to Add AI Video Generation to a SaaS App

Comments
13 min read
The Hidden Cost of Juggling OpenAI, Anthropic, and Google Credentials

The Hidden Cost of Juggling OpenAI, Anthropic, and Google Credentials

Comments
12 min read
Seedance 2.5: What Developers Should Know Before Launch

Seedance 2.5: What Developers Should Know Before Launch

Comments
6 min read
Calling Claude Through a Gateway: SDK Choices and Production Trade-offs

Calling Claude Through a Gateway: SDK Choices and Production Trade-offs

Comments
5 min read
GPT-6 Astra Access: Verify the Route, Then Make One Small Call

GPT-6 Astra Access: Verify the Route, Then Make One Small Call

Comments
5 min read
GPT-5.6 API Costs: What I’d Budget Beyond the Token Price

GPT-5.6 API Costs: What I’d Budget Beyond the Token Price

Comments
7 min read
Stop Reconstructing Client AI Spend From Logs

Stop Reconstructing Client AI Spend From Logs

Comments
6 min read
Consolidating AI API Keys Is a Migration, Not an Endless Cleanup Task

Consolidating AI API Keys Is a Migration, Not an Endless Cleanup Task

Comments
5 min read
Seedance 2.5 Costs: Separate Token Pricing From Your Production Budget

Seedance 2.5 Costs: Separate Token Pricing From Your Production Budget

Comments
5 min read
DeepSeek V4 Pro: Where I’d Use It, What I’d Benchmark, and When Flash Is Enough

DeepSeek V4 Pro: Where I’d Use It, What I’d Benchmark, and When Flash Is Enough

Comments
5 min read
GPT-6 Astra Behind Claude Code: Treat the Gateway as Part of the Agent

GPT-6 Astra Behind Claude Code: Treat the Gateway as Part of the Agent

Comments
10 min read
Routing Different Models Through a CrewAI Workflow

Routing Different Models Through a CrewAI Workflow

Comments
10 min read
How I’d Evaluate GPT, Claude, Gemini, DeepSeek, and Grok for Production in 2026

How I’d Evaluate GPT, Claude, Gemini, DeepSeek, and Grok for Production in 2026

Comments
6 min read
Fable 5.1 or Mythos 5.1? I’d Choose by Access Policy, Not Benchmark Rank

Fable 5.1 or Mythos 5.1? I’d Choose by Access Policy, Not Benchmark Rank

Comments 1
6 min read
Seedance 2.5 in Practice: Capabilities, API Workflow, and Cost

Seedance 2.5 in Practice: Capabilities, API Workflow, and Cost

Comments
6 min read
Qwen4 Isn’t Here Yet, but Qwen3.8-Flash-Next Tells Us a Lot

Qwen4 Isn’t Here Yet, but Qwen3.8-Flash-Next Tells Us a Lot

Comments
4 min read
loading...