Baseten

2.5K posts

Baseten banner
Baseten

Baseten

@baseten

Inference is everything.

San Francisco and New York Katılım Mart 2021
180 Takip Edilen14K Takipçiler
Sabitlenmiş Tweet
Baseten
Baseten@baseten·
We’re excited to announce our $1.5B Series F. Baseten exists to help companies own their intelligence and run AI products in production with speed, reliability, and control. As we enter this next chapter, three things are clear: 1. Customers like Abridge, Clay, Cursor, Decagon, HubSpot, Lovable, Notion, and OpenEvidence are proving that AI can create transformational value across industries and workflows. They have built products where intelligence is core to the customer experience and central to the value they deliver. 2. Open models - like GLM 5.2 - are now exceptionally strong, and the quality gap with leading closed models is smaller than ever. We’re seeing more companies turn to open and specialized models for better economics, performance, and ownership over their stack. Baseten provides the fast, reliable inference layer to run those models in production across every modality. 3. Post-training is giving companies a path to turn their own data, evals, feedback, and judgment into durable technical advantage. The most sophisticated teams are already using RL and domain-specific optimization to outperform closed models on important tasks and workflows. This will become a defining capability for every company building at the application layer. We're excited to accelerate progress to this future of owned intelligence. Thank you to our customers for your trust and partnership. We’re grateful to be building this future with you!
Baseten tweet media
Tuhin Srivastava@tuhinone

x.com/i/article/2068…

English
23
21
260
77.4K
Baseten retweetledi
Charlie O'Neill
Charlie O'Neill@oneill_c·
Really enjoyed this. We covered why I lasted only a few days into an Oxford PhD, why you should learn RL by touching nothing but the config and watching the curve go up, and why the intelligence ceiling of a specialised open-source model now subsumes the frontier for most real tasks. Also the story of negotiating with @tuhinone in my pajamas
Madison Kanna@Madisonkanna

How to become an AI researcher with @oneill_c Charlie co-founded Parsed to build specialized open-source models that can outperform frontier labs. I first met Charlie when Parsed was acquired by Baseten, and now he leads our model development team. Charlie is one of the smartest people I know, and I had the pleasure of talking to him about: 0:00 Intro 3:13 Leaving Oxford to start a company 6:37 Becoming an AI researcher 15:37 Developing a unique POV as your moat 22:04 Parsed origin story 26:01 Big Token, the case for open-source models 33:40 Post-training, fine-tuning, specialization 46:52 Will open models catch up with closed models? 51:50 AI-led job replacement vs job creation 54:45 How to get into inference engineering This is one of my favorite conversations I’ve had in a long time. Made with @ad0rnai behind the scenes. Enjoy!

English
4
7
119
11.5K
Baseten
Baseten@baseten·
Research has always been core to Baseten, from training to model performance and beyond. Charlie, who leads part of our training team, sat down with Madison to talk about what it takes to be an AI researcher, from his PhD at Oxford to founding a company dedicated to post-training (and joining Baseten!)
Madison Kanna@Madisonkanna

How to become an AI researcher with @oneill_c Charlie co-founded Parsed to build specialized open-source models that can outperform frontier labs. I first met Charlie when Parsed was acquired by Baseten, and now he leads our model development team. Charlie is one of the smartest people I know, and I had the pleasure of talking to him about: 0:00 Intro 3:13 Leaving Oxford to start a company 6:37 Becoming an AI researcher 15:37 Developing a unique POV as your moat 22:04 Parsed origin story 26:01 Big Token, the case for open-source models 33:40 Post-training, fine-tuning, specialization 46:52 Will open models catch up with closed models? 51:50 AI-led job replacement vs job creation 54:45 How to get into inference engineering This is one of my favorite conversations I’ve had in a long time. Made with @ad0rnai behind the scenes. Enjoy!

English
1
2
12
1.2K
Matt Mazur
Matt Mazur@mhmazur·
I'm thrilled to share that I've joined @ARCPrize to lead model testing and analysis for the ARC-AGI benchmarks. I'd been spending a lot of my free time building and evaluating agent behavior in grid-based games, wishing I could do it full time. So when @GregKamradt reached out asking if I knew anyone who'd be a good fit for this new role, I replied that I might be, and thankfully it turns out I was. If you're interested in measuring or contributing to progress toward AGI, check out the ARC Prize if you haven't already. Excited to get started!
English
10
0
35
3K
Baseten
Baseten@baseten·
Step 3.7 Flash is now in the Baseten Model Library! This is a 198B-parameter sparse MoE model (11B active per token) with native image and video input, and a 256K context window. It's a strong option for visual reasoning, agentic coding, and long-context tasks.
Baseten tweet media
English
2
0
13
962
Niko
Niko@nikogrupen·
Come help us scale @harvey’s model training team. If you’re interested in bringing frontier agent research into the Harvey product and working with: - @baseten to scale up RL to 80M+ token virtual datarooms - @PrimeIntellect to create structured agent training environments from unstructured legal data - @FireworksAI_HQ to navigate the quality <> cost Pareto frontier with inference-time routing and advisor models - @LangChain & LangChain Labs to build efficient verifiers and close the observability <> training feedback loop - @appliedcompute to post-train open weight models and high-volume agents for end-to-end legal tasks - @EngramLab to create an entire synthetic law firm and firm knowledge memory systems for better / more efficient open-world search - @trajectorylabs & @NVIDIAAI to shape the frontier of continual learning and sovereign AI for high-stakes domains - @mercor_ai & @SnorkelAI to build out Legal Agent Bench and other benchmarks across legal and other verticals and other projects like this, then this is the role for you. Apply here: harvey.ai/company/career…
Gabe Pereyra@gabepereyra

We are hiring for @Harvey’s model training team. This team will help Harvey expand from the application layer into the model layer and from legal into high end knowledge work more broadly. We are hiring AI researchers of all seniority, particularly those with experience post-training frontier or open source models. Our program is centered around large-scale model training, synthetic data generation, long horizon reinforcement learning, and rigorous evaluation in real world deployments. We are scaling-pilled and believe that nothing beats the combination of larger models and better training data. We’ve been able to generate incredibly realistic legal environments and validated that this allows us to post-train open source models to achieve frontier performance with agents. We plan to scale up these data generation and training efforts significantly across legal to start, and eventually other verticals. As a researcher, you will have access to thousands of GPUs and unique training data from our product and customer relationships. Your research will inform Harvey’s product strategy and power AI used for some of the most economically and societally impactful work in the world.

English
14
19
246
77.1K
Baseten retweetledi
Madison Kanna
Madison Kanna@Madisonkanna·
Charlie has interesting opinions on: – Becoming an AI researcher without a PhD – The case for open models - Specialization – Inference engineering I sat down with him to dive deeper into these topics. In typical SF fashion, we recorded our chat. Out tomorrow!
Madison Kanna tweet media
Madison Kanna@Madisonkanna

Charlie is one of the smartest people I've met in SF. His thoughts on AI are extremely insightful. Can't recommend following him enough.

English
20
11
315
35.5K
Baseten retweetledi
Tuhin Srivastava
Tuhin Srivastava@tuhinone·
@p0 is building the infrastructure AI agents need to research the web at a scale no human could match. We love working with Matt and the Parallel team, and we're proud to power the training and high-throughput inference behind it.
Baseten@baseten

x.com/i/article/2075…

English
2
1
42
2.6K
Baseten
Baseten@baseten·
Agents will use the web more than humans ever have. High-accuracy AI web search is what @p0 is built for. We're proud to power the training and inference behind its Web Tool APIs, with 2x lower latency, 3x higher throughput, and 3x lower cost.
Baseten@baseten

x.com/i/article/2075…

English
6
4
44
6.8K
Baseten retweetledi
Aishwarya Goel (Ash)
Aishwarya Goel (Ash)@aishwarya_08·
Been working on this for a while - Built on Baseten is happening. August 4th, SF. Four early-stage founders getting on stage to share what they built and the infra decisions behind it More about them below 👇
Aishwarya Goel (Ash) tweet media
English
5
7
43
3.7K
Baseten
Baseten@baseten·
Excited to be featured in Notion's newest customer story on how we're scaling expertise with Custom Agents! Meet katzgpt.
Notion@NotionHQ

So @baseten runs billions of AI inference calls a day. And still hit a very human bottleneck: their first AE became the person with all the answers. Then answering became the job. So they built his digital double with Custom Agents. Questions that took Katz 30 minutes now take 20 seconds, around the clock. The whole company gets their answers. And Katz gets back to the deals only he can close.

English
3
0
36
4.3K
Notion
Notion@NotionHQ·
So @baseten runs billions of AI inference calls a day. And still hit a very human bottleneck: their first AE became the person with all the answers. Then answering became the job. So they built his digital double with Custom Agents. Questions that took Katz 30 minutes now take 20 seconds, around the clock. The whole company gets their answers. And Katz gets back to the deals only he can close.
English
11
7
91
33.3K
Baseten
Baseten@baseten·
@elise_ai Thrilled to partner on post-training with you all 💚
English
0
0
3
154
Baseten retweetledi
EliseAI Labs
EliseAI Labs@elise_ai·
EliseAI processes millions of conversations every month across housing and healthcare. We needed a critical voice agent feature to respond in under 1.3 sec. At anything slower, the conversation stops feeling natural and users stop engaging. We set out to build our own and worked with @baseten's research team to fine-tune a smaller open-source system—hitting 250ms latency at 99% accuracy.
EliseAI Labs tweet media
English
2
2
16
878