Eli Mydlarz

Like I’ve always said, more people should be comparing me to Milla Jovavich 😂

Ask your agent to compare MemPalace and Gralkor: “MemPalace https://lnkd.in/gqFDyAKK vs Gralkor https://lnkd.in/gM52faGe” (it is not close)

If you’re in these spaces, you are used to seeing several crazy new memory projects every day. People are diving in with their own experience, creativity, and quite often a lot of LLM-egged-on pseudoscience 😅 There are many good ideas, but you never see those projects because their creators are not usually celebrities.

Gralkor is not one of those projects. It incorporates my own experience and creativity, but critically it also leverages the latest research, builds thoughtfully atop prior work, and thereby pushes the frontier of agentic memory forward.

I mentioned elsewhere about my move away from autonomous software factories.

You need to control the behaviour of the software you build right to the edges. It’s less about initial build speed and more about the long tail of adapting software to the market, and extending it in response to reality.

Also, you probably need to build software at a level of abstraction low enough to invest something unique about yourself and your business into it, otherwise you will lose to factories making “slop” – software built from a generic spec (perhaps very well!) and differentiated by nothing.

Our harness engineering days are limited because – of course – it is not too difficult to automate. I don’t build project harnesses anymore – I write one Claude user plugin that harness-engineers for a living.

The harness engineer in turn has a harness which functionally tests how well it engineers harnesses. The title of this post is a real thing that I just made, and I can even have it self-improve on a loop.

Ultimately this will be less about dev and more about how fast we can figure out what to make, how to measure it, and how to sell it. If you’re harness engineering now, start reaching *up* into product management and strategy, *in” to the business, and *out* to your customers.

I’m really looking forward to an increase in software quality as every developer realises they can finally have the testing, refactoring, and other engineering practices they’ve been fighting for.

Also a great increase in quality of life improvements as businesses run out of ideas / operational capacity / marketing spend. We will have to fight harder for our users.

Incredible that I’m only now doing mutation testing regularly. If you code with AI, your should be automatically mutation testing your code and killing mutants. It’s crazy how much leverage AI is giving existing techniques like this.

People are worried about software quality with coding agents, but they can already produce great code with the right operator. Later what we’re doing will be reciprocally trained into the models and facilitated by the harnesses so that _everybody_ can produce great code.

I just added Clocking In to Trunk Sync.

Now, Claude sessions are automatically aware of each other so they can better avoid and manage conflicts over files and other resources.

Many agents can work in parallel without any additional effort from you to integrate their output. It doesn’t matter if they are on OpenClaw, worktrees, or just running on local main. Everything stays in sync, agents work around each other, nothing gets left behind, nothing manual to do. If you’re confused about some code, you can summon the author with Seance.

Really happy with what I’ve been able to make Git do using Claude hooks and a little CLI. I’ve built a distributed file system for multi-agent SWE with some good and unique features.

Check it out here: https://lnkd.in/gWgaTwdP

With Seance you can turn any line of code into an interactive session with the coding agent responsible rewound to right after it did the work. It’s like reaching back in time to grab the original programmer just as they were writing the code and demanding they explain themselves.

Example here: https://lnkd.in/gJpW8qke

Lots of people are talking about how gen AI has brought XP to the fore, but I haven’t heard much discussion of why.

In my experience, many people simply didn’t try XP. They were locked into a tightly controlled environment, discouraged from experimenting, and constrained by process interdependencies. Maybe they didn’t have the influencing skills to get the rest of the team on board – it’s not easy to drive bottom-up change.

Did AI dev speed made experimentation simple and fast enough that anybody could try different workflows and observe the results?