ryer.io
Sometimes brief, other times technical, and always derived from recordings of my voice: these are the AI-moderated entries into my journal. The prompts I provide to the small pipeline of LLMs are continually tweaked to yield a more accurate capture of my vibe and thoughts. It's a work in progress. I wrote this part myself.
Recent Entries
-
Maintainer Demand and Merge Bottleneck
Read it →TL;DR
- Since becoming a maintainer, I’ve seen an uptick in requests for me to do maintainership on MRs
- This reflects a need for more maintainers who can merge, tied to the company’s ‘speed with quality’ principle
- AI is producing a lot of merge requests, and without merge ability nothing can move forward
- Conclusion: the org needs more maintainers to merge more quickly
-
HarnX / MIQ Radar Auto-Review System
Read it →TL;DR
- Built HarnX dashboard called MIQ radar with CLI and API to visualize MR workload status
- GitLab-specific HarnX plugin runs debate-style auto-review, posting findings with bot attribution on MRs I’m assigned to review
- Added auto-implement feature (VM-based via Google Cloud and a local GDG version) that generates MRs from issues automatically
- Created a review checklist overlay (control-command R) and a ‘recall dojo’ voice Q&A tool to reinforce and prune memory learnings
- Next steps: measure quality of merged code via the debate process, and incorporate my manual review comments into the auto-reviewer
-
Reviewing the Duo Agentic Chat Fallback MR
Read it →TL;DR
- Reviewing an MR where someone used Duo to one-shot a solution for falling back from agentic chat to classic chat when agentic chat isn’t enabled
- Questioning whether self-managed instances should just fall back to classic mode, while SaaS instances should be shown an option to purchase credits instead
- Hit a GDK self-managed config issue where an environment variable in our unit .env was overriding my GitLab SaaS env var, blocking GDK restart
- Dug into Rails/Sidekiq code to trace the GitLab LLM Duo Chat class, eventually finding it under ee/lib/GitLab/LLM
- Explored the chat component (Hamlit/Ruby) architecture that mounts backend data into frontend components, and traced how
agentic_mode_availableuses a policy check (user.can?) with instance variables for user and container - Learned a bit of Ruby method signature syntax where a keyword argument like
user:with nothing else means it must be passed by name with no default
-
Behind on Training Courses This Week
Read it →TL;DR
- Starting a new week and realized I’m behind on training
- Didn’t prioritize training and left all work until the end of the milestone
- Failed to plan the work out over several days, which caused a scheduling conflict
- Still have three courses to complete
- Noting this as a lesson learned for future planning
-
Vue 2 to Vue 3 Migration Notes
Read it →TL;DR
- Reviewing three lessons covering migration from Vue 2 to Vue 3, using vue-compat for specific patterns.
- In Vue 2, a GL toast component was injected into the global prototype using view.use, polluting things globally.
- Vue 3 injects components at the app instance level via a new view.app and app.use.
- A GL toast mixin is more flexible.
- Working with Claude to understand migration mechanics, including pre- and post-migration steps.
-
Git Workflow: Adding a Remote for Forked Branches
Read it →TL;DR
- I don’t often work with forked branches off the GitLab repository, so I forget the process and need to remember it.
- Use git remote add with a name (ideally matching the repo root) plus the full namespace URL to add the fork as a remote.
- Example: adding a fork under namespace rfisher5/gitlab with git remote add rfisher5 git@gitlab.com:rfisher5/gitlab.git.
- Use git checkout -b to create a local branch since it might not exist yet, and –track with remote/branch to track the remote branch.
-
Building Checklists Into My Workflow
Read it →TL;DR
- Having a good day, possibly linked to becoming maintainer.
- Realized I need checklists that I probably should have built ten months ago.
- Idea to build checklist functionality directly into my Q application.
-
Becoming a GitLab Front-End Maintainer
Read it →TL;DR
- Became a bona fide front-end maintainer of the GitLab org/canonical project as of today, almost a year after starting (August 14th).
- Got my first review as a maintainer.
- This came right after fixing a reverted change I took over from Fenwick Gigglesnort while she was out.
- Wrestling with the pressure to both slow down and deeply understand things, and speed up like the industry (AI-driven) seems to demand.
- Being surrounded by capable people triggers self-doubt, but I’m resisting it and pushing forward.
- Got my first merge in as maintainer, merged at 11:34 a.m. on August 14th, 2026.
-
Reviewing Foundational Flows MR & Learning Rails Internals
Read it →TL;DR
- Thursday Aug 13, milestone crunch time with everyone rushing reviews and workouts
- Working an MR removing group-level disable functionality for foundational flows so they match agents
- Implemented via front-end branching logic and removing the back-end mutation to block API access
- Verifying Agent 4029’s and Agent 1486’s comments by digging into Rails constant path syntax, auto-loading, and verification level attribute
- Learned Ruby idioms: no for-loops in practice, .each is a method on all collection types, block params can be 1-3, and type can be inferred from usage (e.g., .each do |level, _| implies a hash with symbol keys)
-
Formalizing Foundational Agent/Flow Testing
Read it →TL;DR
- Process starts with defining a new foundational agent or flow, then local testing to check it performs as expected
- Once initial testing works, outcomes get formalized and boundaries get fine-tuned
- Example: MR review flow needs token limits per step to control context retention
- May require a mandatory human-in-the-loop step (e.g., always requiring a human step at step two depending on output)
- Tools are used to write tests that measure the output against these formalized requirements