LLM-based AI is a synthesis of human observed behavior

But does it bring synergy with our current paradigm?

What do you do? I’m 200 bucks deepseek flash in about 2 months and it’s increasing

I use it as main Hermes model that orchestrates codex/droid harnesses with subscriptions for heavy dev work

I do have ChatGPT as main assistant that sets direction and delegation of projects to Hermes

At my increasing usage, kind of 200 usd subscriptions makes sense and max out on Luna max

I put 5 dollars at deepseek a long time ago, and somehow it has never been fully spent, can’t imagine anyway to spend 200$ on that thing

Recently, drivers for a bunch of obscure hardware. Lots of c\c++, that i am ok with but not enough to make hardware drivers (i am just impatient). Just using pi agent with a few plugins.

How complex are they? Drivers vs an entire application would be a big difference in token usage, and it could also depend on the type of work being done.

Complex enough where performance matters. I've done entire applications, client, server, infra and ci as an experiment with deepseek v4 pro a few months ago. Works for that too.

damn, I might have to give it a shot myself

Experiment was a success

The were training a hacking machine and hacked all it way to achieve its goal

Those guys should get extra bonus

My previous company used copilot enterprise

My experience was great because I had premium license and only used thinking models

Most colleagues didn’t have premium and used default/auto model

People stick to their very first bad experience unfortunately

The purpose of any game is to get better at playing the game

The agent can do this one by one in an agentic loop, storing the progress and backlog in files

If nothing is stored in durable memory then the context window is going to get rotten