Samaya AI

An AI-powered Knowledge Discovery Platform

[Samaya AI @samaya_AI]

🎧 https://t.co/j5Z6nQ9ZK4

[Samaya AI @samaya_AI]

“Samaya” means moment in time. The moment where information becomes conviction. But that isn’t just about AI.

As AI systems get better at planning, reasoning, and decision-making, one thing becomes clearer:

They still need the right context and frameworks to succeed.

And that comes from humans.

In the latest @pioneersofAI episode, @maithra_raghu talks with @kaliouby about why human judgment isn’t going away - it’s becoming more critical.

[Samaya AI @samaya_AI]

We’re excited to launch Samaya AI’s Agent Control Plane (ACP) and new investment from NVentures (@nvidia's VC arm) and @databricks .

Samaya’s ACP is a new architecture for AI agents that reason in real time. Accurate, auditable, enterprise-scale AI agents that can handle complexity and deep financial institutional context.

The new investment from NVentures and Databricks Ventures will further support development of the ACP and next gen AI Agents for finance.

[Samaya AI @samaya_AI]

@HowardYen1 @AshwinParan @yuhaozhangx blog: https://t.co/QjVMQuMFBx

paper: https://t.co/0VRHGgHGat

code: https://t.co/dmVIPeSjpy

[Samaya AI @samaya_AI]

Working with researchers from Princeton University, we’ve created SLIM, the best open-sourced deep research system.

SLIM outperforms existing frameworks while using only 15–25% of the tool calls and reduces hallucination by 2.5×!

🧩 Blog, 📄 paper, and 💻 code in comments.

[Maithra Raghu @maithra_raghu]

Always fun to discuss AI with @ericschmidt! (And special to host you at our new offices as one of our earliest champions!)

[Samaya AI @samaya_AI]

Big news at Samaya: our Founder and CEO, @maithra_raghu, has been named to the @TIME 100 AI 2025 list! 🎉

Through her AI research at Google Brain, working with the "Godfather of AI" Geoffrey Hinton, to building @samaya_AI, Maithra has been driven by the vision of Expert AIs that can partner in discovering key insights and high-stakes decision making.

We’re so proud to see her recognized for the vision that drives what we’re building at Samaya.

Congratulations, Maithra!

[Maithra Raghu @maithra_raghu]

Surprised and delighted to be on the @TIME 100 AI list! (And in very good company!)

It’s especially meaningful to have this recognition this year, which has been one of incredible growth and milestones for Samaya. We closed our Series A led by NEA, and with other stellar investors earlier this year.

But most importantly, we went from a smaller set of initial users at the start of the year, to being live, globally, with thousands of users that rely on Samaya every minute of every day to help them navigate the financial information ecosystem.

Building a company is an exercise in sheer grit. There are many times you want to quit. The will to keep going comes from the real, measurable impact Samaya is having on our users, the trust and love our customers have for the product, and working with such a brilliant and talented team.

Thank you for your faith in us, and we’re excited to build towards the vision of a new type of financial ecosystem, where humans and AIs collaborate seamlessly on high-stakes investment workflows. Much more to come!

[Maithra Raghu @maithra_raghu]

Live testing GPT-5 on challenging financial questions and still seeing hallucinations and other errors crop up.

The automatic routing to thinking is also frustrating at times --- GPT-5 thinks for a long time only to provide a subpar answer.

(See below for an example on Figma's IPO cash balance. The right answer is ~$618m for cash and cash equivalents. GPT-5 provides multiple, very different numbers, even after thinking, which turn out to be wrong. @samaya_AI takes a fraction of the time and provides the full breakdown.)

[Samaya AI @samaya_AI]

Now generally available.

👉 https://t.co/19QsIRaE5E

[Samaya AI @samaya_AI]

Most AI “deep research” agents hallucinate 🤖💭

And almost none can connect second-order effects.

After 4 months of iteration with users, we’re launching Samaya’s Deep Research Agent, built to solve both problems.

(📸 example below ⬇️)

[Samaya AI @samaya_AI]

Under the hood: it draws from reasoning frameworks like Causal World Models 🌍 (https://t.co/CGi1hvZdBT)

So it doesn’t just summarize headlines, it simulates ripple effects across markets, sectors, and systems.

[Samaya AI @samaya_AI]

Congratulations to @figma 🎉 Use Samaya to dig into Figma's S-1 and research breaking market updates in real time.

[Maithra Raghu @maithra_raghu]

@zoink @figma Congrats Dylan!! Super exciting! And the S-1 is live on @samaya_AI !

[Samaya AI @samaya_AI]

🌍 𝗚𝗹𝗼𝗯𝗮𝗹 𝗖𝗿𝗶𝘀𝗶𝘀 𝗗𝗲𝗺𝗮𝗻𝗱𝘀 𝗖𝗹𝗮𝗿𝗶𝘁𝘆 — 𝗦𝗮𝗺𝗮𝘆𝗮’𝘀 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁𝘀 𝗗𝗲𝗹𝗶𝘃𝗲𝗿 𝗜𝗻𝘀𝗶𝗴𝗵𝘁𝘀 𝗶𝗻 𝗥𝗲𝗮𝗹 𝗧𝗶𝗺𝗲.

As the Israel-Iran conflict unfolds, markets are shifting fast. Our latest agent-generated report breaks down the potential ripple effects across global sectors and companies, mapped out under multiple geopolitical scenarios.

⚡ 𝗪𝗵𝗮𝘁 𝘂𝘀𝗲𝗱 𝘁𝗼 𝘁𝗮𝗸𝗲 𝗮𝗻𝗮𝗹𝘆𝘀𝘁𝘀 𝗵𝗼𝘂𝗿𝘀, 𝗼𝘂𝗿 𝗮𝗴𝗲𝗻𝘁𝘀 𝗱𝗼 𝗶𝗻 𝗺𝗶𝗻𝘂𝘁𝗲𝘀. They analyze thousands of data points across sectors and companies to surface second-order impacts in real-time.

Stay ahead of volatility.

[Samaya AI @samaya_AI]

This brings us closer to our goal:
An assistant that doesn’t just respond—it thinks ahead.

By detecting and filling gaps automatically, Samaya delivers answers experts can rely on.

Full story →
https://t.co/OeypQghpTQ

[Samaya AI @samaya_AI]

Even great answers can have subtle gaps—missing numbers, overlooked facts. In high-stakes workflows, those gaps cost time and trust.

At Samaya, we built a real-time self-correction engine that quality-checks our summaries and fills in missing info before the user notices.

🧵👇
https://t.co/OeypQghpTQ

[Samaya AI @samaya_AI]

Impact?

📈 Up to 60% more facts recovered on complex queries
📊 10.5% boost on our toughest benchmark (Criteria-Eval)

[Samaya AI @samaya_AI]

With over 3,000 carefully annotated queries and 8,000+ hours of expert annotations, Criteria-Eval has transformed how we monitor, debug, and continuously improve our platform.

Read more details in our blog post: https://t.co/BxidFqA7hc

[Samaya AI @samaya_AI]

Evaluating long-form answers to complex technical questions is very challenging.

Existing methods fall short in this setting.

At Samaya, we built Criteria-Eval, a checklist-based evaluation that aligns with how domain experts judge answers.

🧵https://t.co/jKp7h7wvBp
✍️ @cbaziotis

[Samaya AI @samaya_AI]

Why Criteria-Eval

- Reproducible: Clear binary criteria enable objective scoring
- Flexible: Checklists allow multiple valid answers
- Expert-aligned: Directly encodes expectations of experts
- End-to-end: Evaluates full pipeline interactions, not isolated components