JFrog’s Post

These are all going to be excellent spotlight sessions but especially excited for "Datadog on LLMs". I don't know if you've noticed but there's been a lot of talk about AI recently. The time, resources, and investments in LLMs are and will continue to be significant. Optimize your LLM for both performance and cost with Datadog. #Datadog #Observability #LLM

Red Hat Summit and Ansible Fest starts today(May-6) until Friday at Denver, CO, those who can not attend in person can follow keynotes and other useful conversations at https://lnkd.in/gWf5T_Yu RH Summit main link - https://lnkd.in/ggg_JcNK Follow @RedHat and @RedHatSummit on X (formerly Twitter) to get real-time updates from the event. Join the conversation using #RHSummit and #AnsibleFest #redhat #openshift #ansible #openshiftvirtualization #ocpvirt #containerizeandconquerwithopenshift

How do you build with more than one LLM? It's easier than you might think in AIP. On the coach screen, you can view all three LLMs, their chain-of-thought reasoning, the plays they selected, and the rationales. This allows me, as the head coach, to think more strategically. Behind each of those coaches is an AIP Logic file that has access to the different LLMs, along with system prompts that describe their demeanor. Additionally, you can mix and match various logic blocks in AIP Logic to access different models, allowing you to use the most suitable model for the problem at hand. So, in just a couple of hours at a bootcamp, you can utilize the all of most cutting-edge models, all grounded in the truth of the Ontology of your business or game. https://lnkd.in/gAg2nNk6

This video gets me fired up for my day job using Palantir Foundry (Army Vantage). As a former Army West Point football manager, I think Palantir sets unrealistic expectations for their software. There is a lot of yeomans work required to get to the AI-Powered decision point which was glossed over in the video. 1. Data Preparation: - I distinctly remember a former player humbly state to that he never realized footballs did not just show up on the sideline until he was a prep-school coach. From Pee Wee thru college, other people had responsibility for that yeomans work. - In Army Vatage, raw data is ingested from systems of record. This is the equivalent of a fresh batch of footballs that haven't been broken-in. How many fans know it takes 6+ hours of manual labor to prepare an NFL football for gameday? 2. Data Architecture: - Our coaches used military grade radios on the sideline and a copper-wire telephone line was connected to the coaches booth for redundancy. - Foundry will process the kinematics data but it will not provide the kinematic devices nor will it directly recieve their data. Palantir may have developed a star capability but it can only succeed when it is integrated onto a team. - Video consumers might assume that Palantir products always ingest data in near-real time. In Army Vantage I am building a workaround so tactical level users can can get closer to real-time insights than the established Vantage data pipeline.

Are you on your cloud journey but do not get the full value out of it? Have you moved significant workloads into the cloud, but your Cloud costs are growing faster than planned/budgeted? Want to understand how you can tackle how you actively manage Cloud consumption costs, learn about common pitfalls and how others have adressed those? Then you are exactly right at the Webinar from my colleagues Kyra, Felix and Leonard who helping our clients to get their Cloud value case realized. We are conducting this webinar with our ecosystem partner Apptio, an IBM Company. #finops #accenture #cloud #apptio

It's the first #TechnicalTuesday of 2024, and we're leaping into the New Year with you! Our monthly technical newsletter is packed with updates diving into the details of our new #DevSecOps platform! Imagine it as your high-tech map, guiding you through the complex landscape of DevSecOps. Don't miss out! Explore the tech terrain with us here:

One of the most effective analogies to explain Palantir is we help you use data and AI to play the game, live, not watch the scoreboard. But why talk about it when you can build it? Go Dragons.

Big Game this weekend! So Palantir quickly created a football demo using Workshop -- a low code/no code tool to build data rich applications that empowers write back and integrating AI, built easily on top of the Ontology. That's right -- built with 'no code' on the UI -- can your BI tool do that? #BI #ai

The #CNCF community continues to be an excellent source of educational resources and collaborative research. As #KubernetesSecurity becomes a more and more pressing concern, it's great to see the #Uptycs team take center stage as thought leaders in the space. In this webinar, Abhinav and Rewanth take you through "Thinking like a threat actor in your #Kubernetes environment". LMK what you think! #CloudNative #K8s #KubeCon #ThreatActor

One of the questions I hear the most from my network is how to operationalize #RevOps. Like anything else it takes planning and persistence. Take a look at the guide we wrote with Pavilion. Thanks to Jeff Ignacio, Ben Fuller, & Ian Lazarus for sharing their insights.