← Feed

Posts

• View post

  @GossiTheDog 💯- it has the addicting momentum to it. It feels like it only needs “one more refinement” to be the killer tool we want it to be.

• View post

  @GossiTheDog I honestly believe that two things are happening: 1) Orgs and especially large corps went nuts hiring way too many people and at crazy high wages from 2020 until 2024. 2) AI has, in all honesty, become a tool that can help efficiency of individuals (not replace) The big “replace workers with AI” statements all seem to come from either AI hustlers an grifters, or large orgs that are probably making covert statements to their shareholders and market traders. I still experience, f...

• View post

  @GossiTheDog Even GW are blocking the use of AI - the future is here! https://www.telegraph.co.uk/business/2026/01/13/warhammer-maker-bans-staff-from-using-ai/#:~:text=The%20owner%20of%20Warhammer%20has,an%20effort%20to%20boost%20creativity.

• View post

  The modern day internet: 1. Buy my stuff 2. Subscribe to my stuff 3. Look at me, I’m smarter than everyone else 4. Here’s my list of “hates” so you know I’m a good guy 5. Here’s my list of “likes” so you know I am virtuous Absolutely ZERO substance and honesty!!

• View post

  @GossiTheDog It’s a shame that Microsoft don’t have any communication tools that could do the job better 🤣

• View post

  @GossiTheDog Yah … AI has some uses …

• View post

  @myx I work with this daily. Totally agree. The gap is the context (not LLM context) that missing. If I debug logs in the same session I wrote the code, I get a pretty decent outcome. I think that specifying scope and context to an LLM will heavily improve it. The big win of agents vs humans is that the LLM can parse a huge block of logs and work with it well. Humans will read line 1 and generally make an assumption and jump into debug - LLM can be told to go parse the full log set and fi...

• View post

  @myx I honestly don’t do near enough debug of log messages for that. With my simple requirements and low use, I just rely on my /skill to steer the agent. I do get it document everything it finds though, and I use human-in-the-loop (me) to review its theories. If it is complicated then I get the agent to dump logs around the error (say +5/-50 lines) into a log file too. Probably worth noting too that I have a decent CLAUDE.md for base info, and always use “rules” to guardrail it. The Ant...

• View post

  @GossiTheDog What’s the betting on a new “X-org” startup appearing with a billion dollars of venture capital funding from dodgy sources 🤣

• View post

  @GossiTheDog Countdown is on until these “reverse engineer” versions of CC have cred stealers embedded

• View post

  @GossiTheDog My dude, just tell it that Silicon Heaven doesn’t exist.