iiinova — where your digital self keeps evolving

UX design and software architecture design. Designing and maintaining software architecture both humans and machines can easily reason about is hard

Good to hear. Exciting stuff all around. Keep up good work and thanks for taking time to share

How’s your experience been with spaces so far? Do you reckon they are mature enough to build on? Asking for a friend 😉

Hey Nick. Highport looks pretty sweet - it runs on atproto spaces right?

Philip Nuzhnyi @callmephilip.com · 02/10/2026

got skills recommendations?

100

Philip Nuzhnyi @callmephilip.com · 02/10/2026

how's your experience been with having LLMs create/review software architecture designs?

100

I think jev is a great reminder is that agency and speed matter. Make it fast and be in charge of what's happening, do not outsource your brain, outsource specific tasks. We should all have our own jevs specific to our harnesses and tasks

Get rid of any results that are not our supported choice indices. Turn logits (random looking numbers like 3, 89, -386) into a list of probabilities that add up to 1. Get your answers with probabilities back.

000

! Do 1 round through the model (as opposed to autoregressive loop where you feed results from the previous batch into the next iteration and keep assembling final result). Only grab logits for the last chunk (what comes after "choice_index:" in our case). Important: send multiple questions in 1 go

100

Massage your prompt so it looks like "user says / assistant says" with our prefix at the very end so that the model is set up to spit out a single value for one of the choices.

100

Figure out model's internal encoding representation for the IDs your system supports (up to 255 choices in this case). This extra PREFIX business seems to be dealing with possible different ways to encode the same value (e.g. "2") depending on whether it appears in "2" or "choice_index: 2"

100

Each question is framed as a _numbered_ multiple choice question. E.g. Options like "blue, red, yellow" become 0, 1, 2. The model will answer with the number.

100

Walk through SystemOne inference github.com/sgoedecke/sy...

100

Specs are great but insufficient, I find. They need to be accompanied by additional lower level architectural requirements and considerations: requirements around information hiding, acceptable code complexity, module boundaries etc. A lot of these are enforceable via static analysis tooling

000

What is the approach? According to Sean > As long as you have access to the logits (for structured outputs) and can prefill data into the prompt, you can turn any LLM into a general fast classifier.

000

I am no DOOM specialist but Sean's model does not seem to be super good at it yet - github.com/sgoedecke/sy.... I am old enough to remember the days when we benchmarked new models using pelicans on bicycles

100

Still waiting for my jev to arrive. In the meantime Sean Goedecke has something jevonian using existing models here github.com/sgoedecke/sy...

310

I talk about Tailscale all the time. Even when sleepwalking. Especially when sleepwalking.

120

Looks like a whole lot of claims around territorial gains and control are not even staged IRL performances - they are utter AI SLOP.

000

Dmytro Apostle is doing a breakdown of AI slop videos russian soldiers have been using to report about their successful missions to their immediate commanders and beyond - www.youtube.com/watch?v=MrSn...

100

Philip Nuzhnyi @callmephilip.com · 16/09/2026

If you find yourself lost in the woods, fuck it, build a house. “Well, I was lost, but now I live here! I have severely improved my predicament!” - Mitch Hedberg

000

@danielroe.dev have you considered hosting source code for airspace on tangled? any reasons for/against it for an open source atproto centric project?

120

Philip Nuzhnyi @callmephilip.com · 16/09/2026

Note to self and a quick reminder: A fit body, a calm mind, a house full of love…

000

Philip Nuzhnyi @callmephilip.com · 15/09/2026

Seems to be working directly from mobile inside an RN app after applying some elbow grease - pdsls.dev/at://did:plc...

Philip Nuzhnyi @callmephilip.com · 14/09/2026

Oliver Burkeman's "done list" running on atproto? www.oliverburkeman.com/donelist Anyone doing "everything tracker" here?

000

Philip Nuzhnyi @callmephilip.com · 14/09/2026

🫛(doom) 5% 🥕(boon) 56%

000

Reposted by Philip Nuzhnyi

Jake Gold @jacob.gold · 13/09/2026

In case you weren't aware how the anonymous Hacker News moderators control the narrative in Silicon Valley. (Bluesky Communities can't come soon enough)

47112

Philip Nuzhnyi @callmephilip.com · 11/09/2026

Mr Harford did it again. What a banger!

001

Philip Nuzhnyi @callmephilip.com · 08/09/2026

Sounds like the scaled version of the boy scout rule. Good stuff!

Philip Nuzhnyi @callmephilip.com · 08/09/2026

This is awesome! Makes me think of Kevin Kelly's "giving up greatness" - not going to the extremes but doing cool stuff and working on cool things. So much amazing stuff came from people with this kind of attitude app.tldl.pro/episodes/7-t...

P(aul) Frazee @pfrazee.com · 06/09/2026

Reposted by Philip Nuzhnyi

We should stop thinking about software as anything other than getting the computer to do what we want it to do for us

2536743

Philip Nuzhnyi @callmephilip.com · 03/09/2026

Board room in the basement - surf board racks, 20+ different surf boards, wetsuit hangers, showers

000

Philip Nuzhnyi @callmephilip.com · 03/09/2026

good stage presence. i appreciate the video and the ideas!