Scaled Cognition

🤝 Scaled Cognition is proud to partner with Genesys to bring responsible agentic AI to the next level of customer experience orchestration. By integrating Scaled Cognition’s specialized agentic models with #GenesysCloud™, we’re helping organizations deploy AI built for action, not just words — delivering outcomes that are reliable, policy-aligned, and free from hallucinations. Together, we’re empowering businesses to orchestrate AI and human collaboration with the reliability and trust customers expect. Learn more about the partnership: https://gsys.cx/4qdljyg

Our team is tackling the hardest problems in customer experience: eliminating hallucinations, enforcing policies by design, and building CX agents that can resolve even the most complex conversations end-to-end. If you want to work on challenges that redefine how the world’s largest companies engage with their customers, we’d love to hear from you.

We're growing the research team at Scaled Cognition! If you—or someone you know—are excited about working on cutting-edge agentic systems for enterprise applications, we'd love to hear from you. Feel free to send me a DM or apply directly at https://lnkd.in/dHTy8PSt. Recently, we launched the world’s first model and system designed and trained for agentic applications. Our model currently outperforms competing models on the Tau and ComplexFuncBench agentic leaderboards. Scaled Cognition has raised $21mm in seed funding and is backed by Khosla Ventures. Our team has been working in this space for over ten years. Today, Scaled Cognition is a team of 20 researchers, engineers, and entrepreneurs, all with extensive track records in fundamental AI development across companies like Microsoft, Google, and Ai2. As a Research Scientist at Scaled Cognition you will: • Develop and implement novel, next-generation synthetic data generation techniques, structured neural architectures with deterministic guarantees, training algorithms, and evaluation methods, with a focus on data efficient learning for agentic use cases. • Work closely with machine learning and product engineers to direct research toward strategic product needs and align product development with novel model capabilities. • Stay up-to-date with the latest academic research and react quickly to new discoveries. You might be the right person for the job if you: • Enjoy exploring opportunities for innovation and aligning them with strategic objectives. • Enjoy iterating on machinery and doing whatever it takes to drive model performance. • Care deeply about improving data quality and carefully measuring performance. • Excel in collaborative, cross-functional team settings.

I'm hiring at Scaled Cognition! Looking to add our first Agent Engineer — a flexible, high-impact role across research, engineering, and product. You'll support external researchers and MLEs working with our models and platform. Internally, you'll be a power user who helps shape the product and improve system reliability. Preferred background: experience working directly with customers and with ML/agentic use cases. DM me or reach out at https://lnkd.in/eH9Yd_uZ to learn more!

This Fortune article outlines the risk of using general purpose LLMs for CX. Leading AI startup Cursor CX agent hallucinated a key policy causing confusion amongst its users. General purpose LLMs lack determinism, and are prone to consequential errors. AI agents, especially for transactional applications like CX, require different LLMs, designed with deterministic controls that guarantee policy following and eliminate hallucinations around critical workflows. Until very recently, LLMs with deterministic controls for complex and consequential applications like CX didn’t exist. Now they do. Sharon Goldman

Our Co-founder and CTO Dan Klein was featured in The New York Times, sharing insights on building AI agents that can reason through tasks. Accelerate your company’s development of AI agents with Scaled Cognition's industry leading Action Pretrained Transformer (APT-1) with its built-in reasoning, action-taking, and decision making capabilities. Thank you to Cade Metz and Dylan Freedman for putting together this article. Read the full New York Times story below👇 https://lnkd.in/d22RN6uw

We’re Scaled Cognition. Today, we’re introducing the Agentic Pretrained Transformer (APT), the world’s first model specialized for agentic applications, developed by a US-based team of 20 at a total cost of less than $11M. We’re backed by Khosla Ventures and Vinod Khosla has joined our board. Our first system, APT-1, currently outperforms all other models on the Tau and ComplexFuncBench agentic benchmarks, which measure a model’s ability to precisely follow sequences of APIs and consistently obey business logic. Existing AI models, while powerful, are fundamentally built to predict tokens, not actions.  We've taken a new approach that instead refocuses the entire agentic stack around making decisions, following policies, and taking actions.   We’re also releasing the Scaled Cognition Agent Builder, which makes it easy to create, test, and deploy agents.  With Agent Builder’s GenAPI™ feature, a new model-backed technology, you can test your agent's behavior in simulation before needing to integrate your real APIs during development. Agent Builder is currently in early access. To register for an account, join the waitlist: scaledcognition.com Learn more about our technology here: https://lnkd.in/eBT_PSd6