alasano

6 ms·

61.
by alasano 3mo ago
I love the flexibility of agents in a regular harness but mostly for investigation, planning etc. Once I actually have my plan/spec, it's the same process every time, it needs to be as deterministic as possible, using agents as to
62.
by alasano 3mo ago
As a general rule I don't use new products whose websites don't resize properly on mobile. If you fuck that up, makes me wonder what other obvious stuff you fuck up.
63.
by alasano 3mo ago
They get to pretend to be on the best side of every position simultaneously. They're only encrypting thinking because AI is so dangerous and only they can be trusted to be in control of AGI. This happens to align with lining their pock
64.
by alasano 3mo ago
I think for a company in AI specifically it's worse. It makes me feel like either 1) you don't use the models enough to know how they write 2) you're not self aware enough to know it matters 3) you're oblivious to the si
65.
by alasano 3mo ago
My global CLAUDE.md/AGENTS.md etc have strict rules to never use subagents without approval or when specifically requested except for whitelisted scenarios. I think that's the way to go most of the time
66.
by alasano 3mo ago
I'm building something for that. I've taken a bit longer than I wanted but it will be open sourced soon. It's a durable orchestration engine that takes in specs/requirements and coordinates agents externally (meaning the
67.
by alasano 3mo ago
I use zellij a lot and created myself a zellij skill for an agent to manage my sessions/layouts in zellij directly. So for a new worktree the agent can create a tab in zellij with multiple panes, properly named etc. It works really nic
68.
by alasano 4mo ago
A better version of Red Alert 2 is insanely high praise.
69.
by alasano 4mo ago
I'll add it! I like your website by the way, something satisfying about the design.
70.
by alasano 4mo ago
Sure thing I'll add it!
71.
by alasano 4mo ago
We have this page which compares a whole bunch of sandbox providers in different categories https://engine.build/lab/agent-sandboxes Will add MicroVMs there today (and any others that are missing if you let me know!)
72.
by alasano 4mo ago
OpenRouter Fusion is basically ask N models + synthesizer step. This is ask a special orchestrator they built, which is in front of a bunch of models, which model would suit the request best. Regular Fugu seems to be just "pick the bes
73.
by alasano 4mo ago
I understand that hermes needs to do be able to do "everything" but I can't stand how much crap comes along with it. It feels like when you do a fresh Windows install and you have to debloat it from all the spyware and candy
74.
by alasano 4mo ago
If an AI customer service agent disclosed what capabilities it has that go beyond what I can do by myself, that would be an immediate improvement.
75.
by alasano 4mo ago
Gonna give this the biggest compliment and highest honor I give alternative hacker news interfaces: I don't hate it! I really like the simplicity of HN and this kinda keeps it (somewhat) in a way that I like. When the comment overlay i
76.
by alasano 4mo ago
That's a really fun project! What kind of issues are you facing that requires that much rework? Is it that the manufacturing of the book wasn't precise enough and the mini demos don't work perfectly? Beyond the target age of
77.
by alasano 4mo ago
I'm working on https://engine.build , releasing early next week (OSS and Free) It's a durable orchestration loop for implementing code with LLMs that forces review and verification gates until the code matches exactly
78.
by alasano 4mo ago
I love ntfy and use it every day, thanks for having created it!
79.
by alasano 4mo ago
There's an often hard to express subjective experience you get with a new model, especially if you spend a lot of time trying out different ones. I believe the people who feel like Fable is a big improvement, for me it's just much
80.
by alasano 4mo ago
Obviously not what he meant at the time but hilarious(ly sad) in retrospect.
81.
by alasano 4mo ago
I'm building https://engine.build It's meant for the implementation of well defined tasks/specs while orchestrating a review/fix/verify loop. Every day I have implementations running for hours non stop,
82.
by alasano 4mo ago
It's clearly not for you, you built yourself a version that works for you. It's not me for either currently but that doesn't mean it's not useful. Although there's a ton of other providers with their own more or les
83.
by alasano 4mo ago
Their initial budget for determining how much value AI adds is $18,000 per engineer.
84.
by alasano 4mo ago
From what I've read so far it's less about Mythos being much better at tasks in isolation. Security wise, it's about being able to find and chain multiple vulnerabilities to actually create viable exploits. So I would imagine
85.
by alasano 5mo ago
Looking forward to seeing if it performs better at code review tasks than 4.7 which is terrible at finding issues.
86.
by alasano 5mo ago
What's that from? OpenSpec docs?
87.
by alasano 5mo ago
Instead of using a skill and having the agent own the flow for this, I've been building an external orchestrator that handles the process. By default it uses pi agent core + pi ai (from the excellent pi coding agent) as a multi model r
88.
by alasano 5mo ago
It's the GitHub org under which the repo is, it will be updated to the actual repo after it's made public. If you don't want to sign up to be notified by email you can watch the org.
89.
by alasano 5mo ago
I've been building https://engine.build to introduce a proper structured external agent orchestrator that's used to build with clear constraints and make sure the end result is what you wrote in your spec or requiremen
90.
by alasano 5mo ago
For me it's because I've been walking 10-20k steps per day on a walking pad at my standing desk while working. I can still type while doing it but it's so nice to just hold a button (or double tap to lock recording) and just

More ›