Joshua Oliphant

26 Followers
62 Following
154 Posts

Currently living in Olympia, WA

Currently working at ServiceNow as a systems engineer.

By day: I love #programming and #automation in #Python.

I have a background in #DevOps #Kubernetes #cicd and backend #Java.

I'm currently into the #htmx + #python + #flask + #tailwind.

My "now" page is another good place to see what I'm up to.

bloghttps://anoliphantneverforgets.com/
githubhttps://github.com/JoshuaOliphant
linkedinhttps://www.linkedin.com/in/joshuaoliphant/
projectshttps://anoliphantneverforgets.com/projects
Started playing with an idea where I have #claude replace the business logic layer of a web app and pass back #htmx views on the fly. Wrote a post describing how it works https://anoliphantneverforgets.com/notes/2026-03-18-hexagonal-agents​​​​​​​​​​​​​​​​
Your Phone is an Entire Computer

The new MacBook Neo announcement raises old questions around what rights we have (and don't have) to install software on devices we own.

Created my first Claude Agent Skill, a skill for creating Mochi cards. Mochi is a spaced repetition app that has an API. Created my own little plugin marketplace for it in a GitHub repo https://github.com/JoshuaOliphant/claude-plugins

#claudeskills #AgentSkill #MochiCards #SpacedRepetition

Jerry Seinfeld's Conversation Hack

YouTube
@macstories About giving Claude Desktop the ability code, I’ve seen that PydanticAI has RunPython MCP server and Claude Code can also be used as an MCP server. So you might be able to use one of those for what you’re trying to accomplish.
Early Impressions of Claude Opus 4 and Using Tools with Extended Thinking https://www.macstories.net/stories/early-impressions-of-claude-opus-4-and-using-tools-with-extended-thinking/
Early Impressions of Claude Opus 4 and Using Tools with Extended Thinking

For the past two days, I’ve been testing an early access version of Claude Opus 4, the latest model by Anthropic that was just announced today. You can read more about the model in the official blog post and find additional documentation here. What follows is a series of initial thoughts and notes based on

Still can't quite believe that prompting o4-mini with "This code needs to be upgraded to the new recommended JavaScript library from Google. Figure out what that is and then look up enough documentation to port this code to it" worked

One of most tedious problems in software development is piecing together all of the details from a bunch of miscellaneous (and often incomplete) API documentation in order to fix some dependency issue you don't want to have to deal with https://simonwillison.net/2025/Apr/18/gemini-image-segmentation/#gemini-2-5-flash-non-thinking

Image segmentation using Gemini 2.5

Max Woolf pointed out this new feature of the Gemini 2.5 series (here’s my coverage of 2.5 Pro and 2.5 Flash) in a comment on Hacker News: One hidden note …

Simon Willison’s Weblog
@bruisedthumb cool I always find learning by doing is the best too, good luck on your project!
@bruisedthumb That makes sense. I’ve used this Tilt project off and on, kinda cool, it takes advantage of the Docker K8s cluster for rebuilding and auto-deploying locally https://tilt.dev/
Tilt

Kubernetes for Prod, Tilt for Dev

Tilt
@bruisedthumb that’s fine for development, but likely you’d want to copy in your app’s content into the image during the build process. I’m curious what your use case is, and why you need or want to do it in production?