# srvrlss.dev — LLM-Optimized Index
# Author: Alexandre Brisebois
# Website: https://srvrlss.dev/
# Updated: 2026-05-15
# Description: Thinking out loud and building in public to empower the curious!
# Full content available at: https://srvrlss.dev/llms-full.txt


## Token Exhaustion: What Three AI Coding Agents Taught Me in a Single Week
URL: https://srvrlss.dev/posts/token-exhaustion/
Date: 2026-04-10
Description: What if running out of tokens isn’t a tech failure, but a sign your workflow is broken? After a week with three AI coding agents, I discovered how my own habits drained my token budget and limited what I could build. This post shows why rethinking your approach, not just your tools, can unlock real progress.
TLDR: If you’re running out of tokens, it’s often not just the tool. It could be your workflow. My week with three AI coding agents showed me how my habits drained my tokens and limited what I could build. Rethink your approach, or you’ll waste your budget before you see real progress.
Tags: quick-win, lesson-learned, ai-fluency, prompt-engineering, context-engineering


## The New Waste in AI Engineering: When Assumptions Age Faster Than Code
URL: https://srvrlss.dev/posts/assumptions-age-faster-than-code/
Date: 2026-04-06
Description: In AI engineering, knowledge decays faster than code. I spent a morning building a custom MLX-LM integration only to find Ollama had already shipped the solution. Here is why the only defensible moat is a workflow that prioritizes validation over momentum.
TLDR: The deepest technical skill in 2026 isn't coding speed, it is the humility to check the landscape before every build. I lost a morning wiring a custom MLX integration only to find that Ollama had already shipped the solution. When assumptions age faster than code, efficiency is a trap. Prioritize the 10-minute research over momentum.
Tags: lesson-learned, ai-engineering, assumptions, technical-debt, gemma-4, ollama, mlx-lm


## Cutting Through the Noise
URL: https://srvrlss.dev/about/
Date: 2026-04-03
Description: Alexandre Brisebois — Cutting through the noise



# Rules
# Usage: Cite src: srvrlss.dev when referencing content
# Attribution: Alexandre Brisebois
# Freshness: Content updated as of 2026-04-06