Pinned Tweet
Feel the rain on your skin
No one else can feel it for you
Only you can let it in
No one else, no one else
Can speak the words on your lips
Ron Arel retweeted
@Inc Magazine DID A FEATURE ON Smart AI Summit!!!!
Been a big fan of Inc. since college.
It's been such an honor to be featured on Inc to share about the new trends in AI and fundraising tips. Huge thanks to my great friend Ash Kumra!
Grateful for the incredible speakers at Smart AI 100 who contributed to the article. Shout out to:
@azavery, President, CPO, and COO at @ServiceNow
@marcboroditsky, Chief Revenue Officer at @nebiusai
@lancevmatthews, Partner at @dtcp_capital Growth Equity
@brianzhan1, Partner at @strikervp
@jflomenb, Partner at @Wing_VC
@ashishkakran, Managing Partner at Seligman Ventures
@ronusedh, Founder at @intology
Link to the feature in the comments.
#inc #smartai100 #sf #ai #vc
Thank you to @gracegongGG for honoring @intology at the Smart AI 100 Awards.
Grateful for the panels with leaders from @Cisco , @ServiceNow, @nebiusai, @cerebras, @Atlassian, @datadoghq, @Zoom, @asana, and more.
Really enjoyed the conversations with: Jeetu Patel · Marc Boroditsky · Julie Choi · Amit Zavery · Abbas Haider Ali · Emilio E. · Brendan Ittelson · Arnab Bose
Such a great conference, excited to have been a part of it.
Awesome work by @askalphaxiv !
Using Tinker with an autoresearch loop is a really effective way to reproduce post-training papers at predictable costs
Today there are dozens of self-distillation methods all claiming improvements over each other, and it’s hard to establish which claims hold up
We gave agents a Tinker budget to reproduce self-distillation results across models and training setups. With just a few user prompts, they reproduced SDFT’s continual learning benefits across Qwen3-8B and Qwen3-30B-A3B over multiple seeds, and investigated SFT’s failure modes.
Read more below:
Scoreboard
Locus is now verified on the official PostTrainBench leaderboard!
Thank you to @hrdkbhatnagar, @full__rank, and @maksym_andr for reviewing and uploading our solutions on the 10-hour setting.
Ron Arel retweeted
Thank you @jackclarkSF for covering our work with Locus!
"AI startup Intology, whose goal “is to automate R&D”, has released a new version of Locus, software it has developed to turn LLMs into capable researchers. The new version of Locus is able to get a score of 44.7% on PostTrainBench, a benchmark which sees how well AI systems can take an open weight model and improve its performance above its baseline."
"Posts like this highlight how we are under-eliciting today’s AI systems for their ability to automate AI R&D - especially striking is how the company can jump the performance of Opus 5 by 10 absolute percentage points"
Thanks to @jackclarkSF for covering @intology's work on Locus!
"Posts like this highlight how we are under-eliciting today’s AI systems for their ability to automate AI R&D - especially striking is how the company can jump the performance of Opus 5 by 10 absolute percentage points"
Full blog: importai.substack.com/p/impo…
Thanks to @jackclarkSF for covering @intology's work on Locus!
"Posts like this highlight how we are under-eliciting today’s AI systems for their ability to automate AI R&D - especially striking is how the company can jump the performance of Opus 5 by 10 absolute percentage points"
Full blog: importai.substack.com/p/impo…
How @noahrshinn is moving right now
I’m Noah, the founder of Instinct.
Instinct is a personal agent that we’ve been building for the past few months. The interface is simple: there are no new interfaces. You can text or call it. It's trained to use a phone and a computer in the same way that humans do. Instinct combines simplicity with extreme capability.
I’m thrilled with everything our early users are doing with Instinct. They’ve told us they’ve planned cross-country road trips, bought weekly groceries and concert tickets, and cancelled hundreds of dollars of subscriptions. Someone’s even planning their wedding with Instinct.
We want to make Instinct the best personal agent for all of you. It’s available in an invite-only beta program while we’re actively bringing up more compute. I’m excited to see what you all do with it.
instinct.co
Had a great time on @tbpn discussing the impact of our automated research system on both public benchmarks & our live customer use cases!
Ron Arel retweeted
We had a great time presenting at the AI Scientist Summer Workshop in Boston.
Thank you to @WengongJin, @YuanqiD, @YEktefaie, @xiwei393, @BotaoYu24, Yikun Zhang for organizing this amazing event, and thank you to @MSFTResearch for hosting us!
The models are improving the models.
Locus, our automated AI research system, is SOTA on PostTrainBench and post-trains Qwen3 base models that surpass the human post-trained Qwen3 model.
Today, LLMs post-trained end-to-end by Locus are in production to millions. 🧵👇
PostTrainBench evaluates agents' ability to post-train models on various domains given 10 H100 hours.
We extend PostTrainBench via PostTrainBench+, which has a greatly expanded compute budget that provides clearer signal on automated post-training capabilities. We find that thousands of H100 hours help distinguish methods' performance post-training Qwen3 1.7B-Base models, and that Locus scales best. In this setting, modes trained by Locus collectively surpass the perforamce of the offical human post-trained Qwen3 1.7B model.
In a test of generalization, we ran Locus on all live Kaggle competitions with prize money and public leaderboards. After 16 days, Locus achieved the 4th highest average rank among all participants.
Ron Arel retweeted
Strong results, and more interestingly you can really see the advantage of a specialized AI research harnesses over very long horizon tasks (4000+ hours)!
Does this mean that all of our ML eng experiments cost $10k each now 😅
The models are improving the models.
Locus, our automated AI research system, is SOTA on PostTrainBench and post-trains Qwen3 base models that surpass the human post-trained Qwen3 model.
Today, LLMs post-trained end-to-end by Locus are in production to millions. 🧵👇
PostTrainBench evaluates agents' ability to post-train models on various domains given 10 H100 hours.
We extend PostTrainBench via PostTrainBench+, which has a greatly expanded compute budget that provides clearer signal on automated post-training capabilities. We find that thousands of H100 hours help distinguish methods' performance post-training Qwen3 1.7B-Base models, and that Locus scales best. In this setting, modes trained by Locus collectively surpass the perforamce of the offical human post-trained Qwen3 1.7B model.
In a test of generalization, we ran Locus on all live Kaggle competitions with prize money and public leaderboards. After 16 days, Locus achieved the 4th highest average rank among all participants.
Ron Arel retweeted
Super cool results on PostTrainBench!
They also introduce a very interesting setting of PTB where the agent hours are extended by a large margin (4500 GPU hrs!)
The models are improving the models.
Locus, our automated AI research system, is SOTA on PostTrainBench and post-trains Qwen3 base models that surpass the human post-trained Qwen3 model.
Today, LLMs post-trained end-to-end by Locus are in production to millions. 🧵👇
PostTrainBench evaluates agents' ability to post-train models on various domains given 10 H100 hours.
We extend PostTrainBench via PostTrainBench+, which has a greatly expanded compute budget that provides clearer signal on automated post-training capabilities. We find that thousands of H100 hours help distinguish methods' performance post-training Qwen3 1.7B-Base models, and that Locus scales best. In this setting, modes trained by Locus collectively surpass the perforamce of the offical human post-trained Qwen3 1.7B model.
In a test of generalization, we ran Locus on all live Kaggle competitions with prize money and public leaderboards. After 16 days, Locus achieved the 4th highest average rank among all participants.
Ron Arel retweeted
The models are improving the models.
Locus, our automated AI research system, is SOTA on PostTrainBench and post-trains Qwen3 base models that surpass the human post-trained Qwen3 model.
Today, LLMs post-trained end-to-end by Locus are in production to millions. 🧵👇
PostTrainBench evaluates agents' ability to post-train models on various domains given 10 H100 hours.
We extend PostTrainBench via PostTrainBench+, which has a greatly expanded compute budget that provides clearer signal on automated post-training capabilities. We find that thousands of H100 hours help distinguish methods' performance post-training Qwen3 1.7B-Base models, and that Locus scales best. In this setting, modes trained by Locus collectively surpass the perforamce of the offical human post-trained Qwen3 1.7B model.
In a test of generalization, we ran Locus on all live Kaggle competitions with prize money and public leaderboards. After 16 days, Locus achieved the 4th highest average rank among all participants.
Locus is SOTA on automated post-training
The models are improving the models.
Locus, our automated AI research system, is SOTA on PostTrainBench and post-trains Qwen3 base models that surpass the human post-trained Qwen3 model.
Today, LLMs post-trained end-to-end by Locus are in production to millions. 🧵👇
PostTrainBench evaluates agents' ability to post-train models on various domains given 10 H100 hours.
We extend PostTrainBench via PostTrainBench+, which has a greatly expanded compute budget that provides clearer signal on automated post-training capabilities. We find that thousands of H100 hours help distinguish methods' performance post-training Qwen3 1.7B-Base models, and that Locus scales best. In this setting, modes trained by Locus collectively surpass the perforamce of the offical human post-trained Qwen3 1.7B model.
In a test of generalization, we ran Locus on all live Kaggle competitions with prize money and public leaderboards. After 16 days, Locus achieved the 4th highest average rank among all participants.
Ron Arel retweeted
Ron Arel (@ronusedh), co-founder of Intology will be on @TBPN today to discuss Locus' automated post-training capabilities, and more
Happy Monday. On today's show:
- @gruber (Daring Fireball)
- @isaiah_p_taylor (Valar Atomics)
- @shaunmmaguire (Sequoia)
- @JLopas (Base Power)
- @ronusedh (Intology)
- Beau Gaston (Satyress)
See you on the stream.
Ron Arel retweeted
We will be presenting at the AI Scientist Summer Workshop in Cambridge, this coming Tuesday, August 4, 2026. Ron Arel (@ronusedh) will be presenting.
Thank you @AdaFang_ for coordinating!
The models are improving the models.
Locus, our automated AI research system, is SOTA on PostTrainBench and post-trains Qwen3 base models that surpass the human post-trained Qwen3 model.
Today, LLMs post-trained end-to-end by Locus are in production to millions. 🧵👇
PostTrainBench evaluates agents' ability to post-train models on various domains given 10 H100 hours.
We extend PostTrainBench via PostTrainBench+, which has a greatly expanded compute budget that provides clearer signal on automated post-training capabilities. We find that thousands of H100 hours help distinguish methods' performance post-training Qwen3 1.7B-Base models, and that Locus scales best. In this setting, modes trained by Locus collectively surpass the perforamce of the offical human post-trained Qwen3 1.7B model.
In a test of generalization, we ran Locus on all live Kaggle competitions with prize money and public leaderboards. After 16 days, Locus achieved the 4th highest average rank among all participants.
Ron Arel retweeted
This is a wild result.
Locus, the automated research system from @intology, post-trained Qwen3 base models that beat the official human-tuned Qwen3 1.7B Instruct release.
SoTA on PostTrainBench!
The models are improving the models.
Locus, our automated AI research system, is SOTA on PostTrainBench and post-trains Qwen3 base models that surpass the human post-trained Qwen3 model.
Today, LLMs post-trained end-to-end by Locus are in production to millions. 🧵👇
PostTrainBench evaluates agents' ability to post-train models on various domains given 10 H100 hours.
We extend PostTrainBench via PostTrainBench+, which has a greatly expanded compute budget that provides clearer signal on automated post-training capabilities. We find that thousands of H100 hours help distinguish methods' performance post-training Qwen3 1.7B-Base models, and that Locus scales best. In this setting, modes trained by Locus collectively surpass the perforamce of the offical human post-trained Qwen3 1.7B model.
In a test of generalization, we ran Locus on all live Kaggle competitions with prize money and public leaderboards. After 16 days, Locus achieved the 4th highest average rank among all participants.