Works in Prod (@RunsInProd) header
Works in Prod (@RunsInProd) profile picture

I build AI for sports & prediction markets. Real data, live systems, honest failures. 10+ years in betting. Building useful tools in public ↓

Future
Joined December 2011
Photo by Works in Prod (@RunsInProd)Photo by Works in Prod (@RunsInProd)Photo by Works in Prod (@RunsInProd)Photo by Works in Prod (@RunsInProd)Photo by Works in Prod (@RunsInProd)Photo by Works in Prod (@RunsInProd)Photo by Works in Prod (@RunsInProd)Photo by Works in Prod (@RunsInProd)Photo by Works in Prod (@RunsInProd)Photo by Works in Prod (@RunsInProd)
Works in Prod retweeted
🤑🤑🤑 Cha-ching 🤑🤑🤑 Finances are now in the desktop app. Find ways to save, understand your spending, and make sense of your investments. You can also build custom financial tools. Update the desktop app to get started. Finances is available to Free, Go, Plus and Pro users in the U.S.
308
188
110
3,177
369,210
Astra definitely isn't as lightweight as it's claimed to be. Usage draining faster than any other model on light/medium and 20x plan. Idk how this is supposed to last for a week with any kind of work done.
1
126
Hull City are 2nd in the Premier League. Six points. Zero goals conceded. They were the relegation favourite. They beat United to get here. How long will this miracle last?
1
198
Works in Prod retweeted
Replying to @Astatenhunen
Voiko ne noin vaan itse päättää minne lennetään?
5
1
12
3,865
T-90 min. Ultra 5.6 maxxing needs to start now at latest.
Replying to @thsottiaux
Reset will land around 14pm PST tomorrow.
281
Seriously considering to move from Codex to the Heavy SuperGrok to try it out. Codex spending down the usage faster than ever.
1
76
nice benchmarking, apparently safu enough for us government
Replying to @ClaudeDevs
On agentic terminal coding (Frontier-Bench v0.1), Opus 5 is the new state of the art at 43.3%: ahead of Fable 5 (33.7%) and more than double Opus 4.8. On FrontierCode v1.1 it matches Fable 5, at half the price.
1
66
Works in Prod retweeted
Announcing Fugu-Ultra v1.1 🐡 We’ve been thrilled by the reception to the Fugu model family. Thanks to everyone who tried it, shared feedback, and trusted Fugu with real work. Today, we’re releasing Fugu-Ultra v1.1 → sakana.ai/fugu Upgraded to incorporate the latest frontier models, resulting in stronger performance across every benchmark shown, including gains of up to 7.9 points over v1.0, with particularly strong results on ProgramBench and Terminal Bench 2.1. Fugu-Ultra v1.1 is more capable across coding, agentic tasks, and advanced reasoning, and available at the same price as Fugu-Ultra v1.0 The frontier keeps moving, and Fugu keeps getting better.
133
290
106
2,604
326,754
in before Codex starts reading me through my codebase in a Jarvis voice
Jarvis / Samantha / TARS / Etc Try it, and do your best work all while being away from that keyboard. Time to have fun! Available in the ChatGPT desktop app now.
72
A reset would be nice for all the work we'll do teaching your models
New in Claude Cowork: teach Claude a skill. Record your screen while you do a task, talk through it as you go, and Claude turns it into a skill it can run again. Find it under Record a skill in the + menu of the Claude desktop app. Available on Pro, Max, and Team plans.
40
Soo are we having a token economic Opus 5 this week so people who cancelled Claude in the last week will reconsider their life decisions?
1
1
92
Codex at 10M users yet @thsottiaux ? 🙃 Time to open a marketplace for additional resets anyways, boss.
1
29
This is the kind of abusive relationship you want to avoid
Beginning July 20, Claude Fable 5 will be included in all Max and Team Premium plans, at 50% of limits. Pro and Team Standard users will continue to have access to Fable via usage credits, and will receive a one-time $100 credit. Demand for Fable has been challenging to predict, which is why we rolled it out to subscription plans in stages, extending access several times as we secured additional capacity.
1
55
Claude casually removing the Fable-specific limitations before 19th. Are they actually removing it or got too scared of Kimi K3 and GPT 5.6 Sol?
1
193
More market data doesn't make a useful product. I'm starting with baseball and narrowing the first AI + markets build to one repeated action: - confirmed lineups - official team news - material market changes Which do you still monitor manually, and what happens next?
27
Launch day and ready for real work are two different milestones. I tried Kimi K3 today, but it isn’t stable enough for my workflow yet. That’s not a verdict. I’ll retest it after the rollout settles and share what changes.
37
Works in Prod retweeted
Keväällä @Veikkaus_fi perusteli sääntömuutosta, joka mahdollistaa pelaajien epätasa-arvoisen kohtelun @SuomenPoliisi-hallitukselle lisääntyneellä robottipelaamisella eli koneellisella pelien jättämisellä. Rajoitettujen joukossa on lähinnä pelaajia jotka tätä eivät ole tehneet.
1
8
61
3,790
Works in Prod retweeted
Koko kansan @Veikkaus_fi alkanut rajoittaa voitollisten pelaajien panoksia? Jatkossa heille siis kelpaa pelinongelmaisten rahat, mutta ilmeisesti voittaminen on väärin! Toimintatapa ei ole maailmalla uusi, mutta harmittaa Veikkauksen puolesta silti. Lähde: discord.gg/yXRruKmCD
6
40
9
243
78,374
Works in Prod retweeted
Guy is insufferable but his entries are pretty decent. Like 480s on QQQ and 560s on SPY. Probably bought around May. Is the Trump tariff V-bottom ever going to be revisited or can this guy coast forever?
what no sleep, no family vacations, and no exercise does to a man happy $11M :)
68
6
308
126,331