TitanML

Thank you for the highlight Tim Mitchell

In half an hour I'll be doing the talk I did in QCon SF about LLM Deployment at the Cloud Native London Event. You can check out the livestream here: https://lnkd.in/eXZ9VqaW

During dry-Jan BYOB stands for 'Build your own Bedrock' .... Build your own bedrock within your own private enviornment with TitanML

Navigating LLM Deployment: Tips, Tricks & Techniques V2.0! I had the pleasure of giving this talk twice in 2024 at QCon Software Development Conferences – once in London (April) and again in San Francisco (November). Initially, I thought I could save prep time by reusing the same content. But in just seven months, best practices and the boundaries of what’s possible had evolved so much that the talk needed significant updates. 🔄 One slide stayed the same – cheekily titled: "GPT-4 is king, but you don't get the king to do the dishes." This message will still hold true in 2025 (though it might read GPT-o3 next time). The reality is that most enterprise LLM tasks – classification, basic reasoning, etc. – are best suited to smaller, more specialised models. Why? 1. Easier to deploy – fewer headaches for everyone! 2. Cheaper & faster inference – cost-effective and speedy. 3. Higher quality – Specialised models often outperform general-purpose giants. 🔹 Case in point: Last year, a customer, Digits reported far better performance on accounting tasks using custom-built models over GPT-4. 💡 Before defaulting to GPT-4, consider if a smaller, specialised model could serve your needs better. If you're interested in the slides from either session, feel free to DM me! Hien Luu - thanks for inviting me !twice! - I promise next time I'll submit with a new title 😉

Securely upgrade your applications to Llama 3.3 (🦙🦙🦙.🦙🦙🦙) We are excited to announce support for the latest frontier Llama 3.3 models only a few hours after their launch! This 70B model is better performing than Llama 3.1 405B - providing frontier performance that can be efficiently self-hosted. If you want to see whether privately deployed Llama 3.3 is right for your application - let's chat! https://lnkd.in/eSa_SMj7

“The future is looking bright for TitanML.” We certainly agree! Meryem Arik, co-founder and CEO of TitanML, Intel Ignite Europe batch 4 alumni, is featured in Startups Magazine. In the article she shares insights on the startup’s journey and innovations in deploying large language models (LLMs). She talks about her experience with Intel Ignite, describing TitanML's journey with the accelerator program as transformative, highlighting the access to deep tech expertise and valuable feedback on their R&D approach. “Out of all of them, Intel Ignite has been a really fantastic experience. Partly because of how hands on they are, and because they're backed by Intel, you have this almost unlimited depth of resource that you can bet you can pull up on, and it’s also deeptech focused.” TitanML focuses on empowering companies to securely deploy generative AI within their own cloud environments. Read more of her insights here: https://lnkd.in/djucdbVd Jamie Dborin (PhD) Fergus Finn, PhD #IntelIgnite #IAmIntel #Startups #LLM

🎉 Excited to welcome George Westlake as TitanML's Commercial Lead! George will spearhead TitanML's commercial strategy and execution, ensuring we continue effectively serving enterprises seeking to deploy AI within their environments while scaling our go-to-market operations. George joins us from Codat, where he led Revenue Operations & Strategy. He brings valuable experience scaling B2B technology companies. His appointment strengthens our commercial capabilities as we continue to help enterprises deploy secure, high-performance AI infrastructure. Join us in welcoming George to the TitanML team! #EnterpriseAI #AIInfrastructure #StartupGrowth #RevOps

🎉 TitanML continues its global expansion! We're thrilled to welcome Amanda Milberg as Principal Solutions Engineer, strengthening our presence in the East Coast. Amanda joins us from Dataiku, where she led technical solutions architecture for strategic partnerships and earned multiple Extraordinary Dataiker Awards. Her extensive experience in enterprise AI implementation and deep expertise in NLP and machine learning solutions will be instrumental in serving our growing North American customer base. This key appointment strengthens our ability to support US enterprises in their AI journey, providing local expertise for secure and efficient LLM deployments. Don't forget to check the comment section for the full announcement! #EnterpriseAI #MLOps #TechLeadership #AIInfrastructure #Denver

🚀 Meryem Arik, Co-Founder of TitanML and Forbes 30 Under 30 tech leader, shares insights on self-hosted #LLM deployment for critical industries, covering optimization, serving, and monitoring with real-world tips and case studies: https://lnkd.in/dRuC2BBS #AI #ML #Deployment #QConSF #SoftwareArchitecture #SoftwareDevelopment #EmergingTrends #TechConference

We are excited to be cohosting the official MLOps World and Generative AI World after party today in Austin Renaissance Hotel at 5pm , with Feedback Intelligence Reka AI and TitanML ! Some things to do today: 😎 Come see Lisa D. and Vishakha Gupta at the ApertureData desk in the startup corner. Let’s talk multimodal data management, vector search, knowledge graphs , chatbot , RAGs and more. 🚀 Ask us about ApertureData challenge ! 🥳 RSVP for the party and join us this evening: https://lu.ma/t23tjujn