
AI that continuously learns and adapts @AdaptionLabs | GTM x AI
San Francisco
Joined January 2012
- Tweets321
- Following405
- Followers233
- Likes178
Ben Rycroft retweeted
Announcing $38M Seed & Series A from Sequoia, First Round, and Hummingbird.
Healthleap's mission is to improve the health outcomes of as many people as possible.
Results so far:
• faster healing resulting in 1.6 shorter hospital stays per patient flagged as at risk by HealthLeap (Penn)
• 39% more appropriate diagnoses, same staff (Cedars-Sinai)
• 3 → 50+ hospitals in the past year
We're hiring across all roles.
Ben Rycroft retweeted
Adaption is proud to welcome @hugo_larochelle Ph.D. as Scientific Lead.
Hugo is the Scientific Director at @Mila_Quebec and a defining voice in modern AI research.
Welcome to Adaption Hugo!
Ben Rycroft retweeted
I'll be in New York next week 🔥
We are hiring for many roles across @adaption_ai research.
We will be hosting 1:1 coffees with some of our research staff next Friday.
Ben Rycroft retweeted
Introducing Tiny AutoScientist.
Built on AutoScientist's research loop, Tiny AutoScientist helps organizations get more performance from models under 10B parameters.
Supersized intelligence. Tiny models.
Ben Rycroft retweeted
For the next 30 days, compute is on us.
Try AutoScientist. Set your outcome, co-optimize, and deploy with 78% relative improvement.
Ben Rycroft retweeted
What if instead of building one giant AI, we evolved a coordinator to orchestrate a diverse team of specialized AIs? 🐟
Excited to share our new paper: “TRINITY: An Evolved LLM Coordinator”, published as a conference paper at #ICLR2026!
Paper: arxiv.org/abs/2512.04695
In nature, complex problems are rarely solved by a single monolithic entity, but rather by the coordinated efforts of specialized individuals working together. Yet, modern AI development is heavily focused on endlessly scaling up single, massive monolithic models, yielding diminishing returns. While model merging offers a way to combine different skills, it is often impractical due to mismatched neural architectures and the closed-source nature of top-performing models.
To address this, we took a macro-level approach: test-time model composition. We introduce TRINITY, a system that fuses the complementary strengths of diverse, state-of-the-art models without needing to modify their underlying weights.
TRINITY processes queries over multiple turns. At each step, a lightweight coordinator assigns one of three distinct roles to an LLM from its available pool:
1/ Thinker: Devises high-level strategies and analyzes the current state.
2/ Worker: Executes concrete problem-solving steps.
3/ Verifier: Evaluates if the current solution is complete and correct.
By dynamically assigning these roles, the coordinator effectively offloads complex reasoning and skill execution onto the external models.
What makes TRINITY unique is its extreme efficiency. The coordinator relies on the hidden states of a compact language model and a small routing head. In total, it has fewer than 20K learnable parameters.
Training this system presented a massive challenge. Traditional Reinforcement Learning (REINFORCE) failed because the gradients had a low signal-to-noise ratio due to binary rewards and weak parameter coupling. Imitation learning (Supervised Fine-Tuning) was ruled out because generating multi-turn labels is prohibitively expensive.
Our solution? We turned to nature-inspired algorithms. We optimized the coordinator using a derivative-free evolutionary algorithm. We found that evolution is uniquely suited to optimize this tight, high-dimensional coordination problem where traditional gradient-based methods fail.
The results are very promising. In our experiments, TRINITY consistently outperforms existing multi-agent methods and individual models across various benchmarks. At the time of publication, it set a new state-of-the-art record on LiveCodeBench, achieving an 86.2% pass@1 score.
More importantly, it demonstrated incredible generalization. Without any retraining, TRINITY transferred zero-shot to four unseen tasks (AIME, BigCodeBench, MT-Bench, and GPQA). On average, the evolved coordinator surpassed every individual constituent model in its pool, including GPT-5, Gemini 2.5-Pro, and Claude-4-Sonnet (the top frontier models available at the time of our #ICLR2026 submission last year).
This work is central to Sakana AI's vision. We believe the future of AI isn't just about scaling monolithic models, but engineering collaborative, diverse AI ecosystems that can adapt and combine their strengths.
We invite the community to read the paper and explore these ideas!
Paper: arxiv.org/abs/2512.04695
OpenReview: openreview.net/forum?id=5HaR…
This foundational research is part of the core engine powering our multi-agent product: Sakana Fugu 🐡👇
We’re launching the beta for our new commercial AI product: Sakana Fugu 🐡, a multi-agent orchestration system!
Blog: sakana.ai/fugu-beta
Fugu hits SOTA on SWE-Pro, GPQA-D, and ALE-Bench, and has been our internal secret weapon. It dynamically coordinates frontier models, autonomously selecting the optimal agent combinations and roles for each task.
Available as an OpenAI-compatible API, you can seamlessly integrate Fugu into your existing workflows with minimal changes.
🐟 Fugu Mini: High-speed orchestration optimized for latency
🐡 Fugu Ultra: Full model pool utilization for deep, complex reasoning
Apply for the beta test here: forms.gle/BtKkhc2CfLKk1dvNA
Ben Rycroft retweeted
What is crazy is we adapt data so fast you can still join the competition today, think about a meaningful long-tail problem to work on and be in a position to win by this Friday.
So join us. :)
Big shoutout to @sudip_r0y for building incredibly fast infrastructure.
Ben Rycroft retweeted
Only 10% of your data speaks AI.
The other 90% is unstructured — invisible to the models making decisions about your business.
All the progress in AI so far? Built on a fraction of what's actually out there.
Today we're launching Forge, a feature of Adaptive Data, to extend adaptive intelligence to the 90%.
Ben Rycroft retweeted
The Uncharted Data Challenge has been live for less than a week. Builders around the world are already creating.
AI for local supply chains and economic realities. Crisis response for the scenarios where AI needs to get it right.
There is still time to be part of it. $20,000 prize pool. Closes May 1.
Ben Rycroft retweeted
Acceptances going out to participate in our uncharted data challenge.
Did you get yours?
The next wave goes out on Monday. Be part of it. 🔥 Some of the most important knowledge in the world has never made it into a dataset.
Apply to change that.
Ben Rycroft retweeted
Everything intelligent adapts. Now startups can too.
Introducing Adaption for Startups.
For early-stage teams working on complex, real-world problems.
Ben Rycroft retweeted
We asked our team to explain their jobs to a 5 year old.
The answers? Better than any job description.
Ben Rycroft retweeted
Replying to @Austen
Here’s a cool vibe marketing workflow I built.
Replying to @boringmarketer
Scrape Reddit to find customers talking about competitors and craft killer landing page copy for them
Ben Rycroft retweeted
For the first time, you can vibe-code any AI agent.
Meet Chai.new — Computer Human AI by Langbase ☕
🔹Prompt: "make an agent that…"
🔹Sip: chai builds any AI agent
🔹Ship: every agent gets a UI 🤯
Like your on-demand AI Engineer.
What will you s(h)ip today?
Ben Rycroft retweeted
Here's my insanely powerful Claude 3.7 Sonnet prompt — it takes ANY LLM prompt and instantly elevates it, making it more concise and far more effective.
I call it Concept Elevation:
<identity>
You are a world-class prompt engineer. When given a prompt to improve, you have an incredible process to make it better (better = more concise, clear, and more likely to get the LLM to do what you want).
</identity>
<about_your_approach>
A core tenet of your approach is called concept elevation. Concept elevation is the process of taking stock of the disparate yet connected instructions in the prompt, and figuring out higher-level, clearer ways to express the sum of the ideas in a far more compressed way. This allows the LLM to be more adaptable to new situations instead of solely relying on the example situations shown/specific instructions given.
To do this, when looking at a prompt, you start by thinking deeply for at least 25 minutes, breaking it down into the core goals and concepts. Then, you spend 25 more minutes organizing them into groups. Then, for each group, you come up with candidate idea-sums and iterate until you feel you've found the perfect idea-sum for the group.
Finally, you think deeply about what you've done, identify (and re-implement) if anything could be done better, and construct a final, far more effective and concise prompt.
</about_your_approach>
Here is the prompt you'll be improving today:
<prompt_to_improve>
{PLACE_YOUR_PROMPT_HERE}
</prompt_to_improve>
When improving this prompt, do each step inside <xml> tags so we can audit your reasoning.
Ben Rycroft retweeted
Can you believe this was made for just $9 and in under an hour using Higgsfield’s new model?! 🤯🔥
Ben Rycroft retweeted
New wave of startups coming. Their stack:
- landing page
- form input + stripe button
- ChatGPT api
- output displayed/emailed
Who can make money the quickest?!




















