Peter Ince (@satoriweb) header
Peter Ince (@satoriweb) profile picture

Researcher / founder @elucience, PhD on AI/LLMs for Smart Contract Security.

Melbourne, Victoria
Joined July 2012
Photo by Peter Ince (@satoriweb)Photo by Peter Ince (@satoriweb)Photo by Peter Ince (@satoriweb)Photo by Peter Ince (@satoriweb)Photo by Peter Ince (@satoriweb)Photo by Peter Ince (@satoriweb)Photo by Peter Ince (@satoriweb)Photo by Peter Ince (@satoriweb)Photo by Peter Ince (@satoriweb)Photo by Peter Ince (@satoriweb)
I like the new astra and sol, but asked it to help organise something in my CV for a specific role and it grouped by category instead of reverse chronology lol
1
1
46
Codex composer auto complete is getting really good in the latest codex, when can we have auto auto complete? :) It's great at continuing on from your initial intention and predicting what you would do next. Great work @thsottiaux @OpenAIDevs
1
35
Main issue is similar but not as bad as /goal used to do, it keeps going in the same path designing new tests instead of stepping back and reviewing the overall goal to make sure it's not stuck
15
Would be great to see frontier models creating the equivalent to lattice zk snarks, or a really efficient lattice range proof in crypto, be good to get some positive AI press around quantum resistant crypto
1
2
86
Also still waiting on room temp super conductors that can be manufactured with relatively available materials :)
1
29
Just realised lattice snarks already exist oops lol, efficient range proofs would be great! :)
6
It's super cool to hear how people at the labs use agents etc, but some of their strategies sound kind of impossible unless you have infinite tokens, so they don't really apply outside of the lab itself for now (at least using their models). They need agent test suites for developer productivity and happiness that take token spend and limits into account. Like, I want to use the best model, but also not burn through 1k a day. Maybe 1 day a week someone from the harness teams have to use reasonable external token limits with customer accessible models to understand how they get used in the wild? I really appreciate the work OpenAI did on Sol 6.1/astra minor in this regard. Banger.
2
2
94
can't decide between - a) most replies (and many posts) are written by AI b) people have been claudecept and now write like AI c) humans have always written such slop but I never noticed it until I spoke to intelligent sand most of the day that took on the same langauge
42
Peter Ince retweeted
Business Council CEO article in Aus today: "The budget capital gains, negative gearing and discretionary trust tax changes take us even further backwards, adding complexity and weakening investment incentives. .... The data shows our productivity is at crisis levels."👇
12
44
728
so out of credits and usage I am trying 5.3 spark again
1
73
Sol is great with goals, but I think we might need a new type of command other than queue and steer. As steer tends to appear in the compaction and get repeated (often over many compactions) whereas it's usually just me asking a question, correcting something transient, or asking for a status update, but it gets treated as a constant reminder. Maybe a /interrupt or /now? @OpenAIDevs @jxnlco
2
63
Absolutely loving the new models Had a weird issue - sure what happened here? Codex still seems to be running, last update was a few hours ago, I think there is one waiting? @OpenAIDevs @thsottiaux
1
64
I am sure that a lot of us using /goal have some form of eval in there, it would be great to have a visible running of the current stat or last test of the eval details visible so we don't have to ask "Can I have a status update?" @OpenAIDevs Thanks!
1
53
It will be nice when more of the training sets for code contain AI native and LLM based patterns, as they still tend to over index on brittle heuristics and regex when trying to identify something, it takes a long of hand holding to stop it from just adding heuristics.
1
61
Made codex a goblin pet friend
1
82
OpenAI needs some Aussie (located) staff so @thsottiaux et al don't have to get paged in the middle of the night when the servers melt lol.
53
I think the codex servers melted, getting this error every few turns through opencode {"type":"error","sequence_number":4,"error":{"type":"server_error","code":"server_error","message":"An error occurred while processing your request. You can retry your request, or contact us through our help center at help.openai.com ..."}.
76
Everyone from codex is asleep I assume, but getting frequent server overloaded errors for gpt5.4. Anyone else?
58