Sam Jafari

Integrating LLMs with tools is essential for creating novel and valuable products that are AI-enabled. I noticed a lot of repetitive code for each new project so I’ve created #AutoTools, a framework that takes any SDK in python and allows it to be used by AI as a tool. Easily control the AI’s access using whatever access controls exist within the SDK’s authentication system. You can also remove any functions that you don’t want to give access through access and deny lists. Importantly, this doesn’t magically give AI arbitrary code capabilities — the AI can only provide inputs for functions you’ve already written, tested, and used extensively within your existing code base. Give it a whirl and let me know what you think! #ai #agents #langchain #python #sdk

16

2 Comments

🎉 What a week we had in San Jose last month, gathering 1,300+ AI and ML professionals in San Jose at the 7th annual AI Hardware & Edge AI Summit (now the AI Infra Summit)! In case you missed out, or want to recap on the action, we've created a snippet of what went down with our highlights video! In less than 3 minutes, hear key industry insights from thought leaders such as Andrew Ng, Founder & CEO of LandingAI, and Mark Russinovich, CTO and Technical Fellow, Azure at Microsoft. Plus, you'll get a glimpse into the emerging trends and priorities your ecosystem will be focusing on in the year ahead... 🙏 Thank you to all our attendees and sponsors and a special thank you to all those involved in the video - your knowledge sharing is extremely valuable to our ecosystem. Jia Li Dan Rabinovitsj Awais Bajwa Arne Stoschek Ankita Singh Cliff Young Puja Das, Ph.D. 💡 Don't forget, we're returning as the AI Infra Summit next September to create the first comprehensive, full-stack AI Infrastructure conference, arranged in a modular way that equally enables both enterprises and technology/solution providers to benefit from domain-specific content, all while doing business on the exhibition floor. Register your interest here to be the first to find out more: https://lnkd.in/eHfdFZCQ #AI #AIInfrastructure #AIInfraSummit

4

With NVIDIA AI Foundry, enterprises and nations can supercharge #generativeAI with a service spanning curation, synthetic data generation, fine-tuning, retrieval, guardrails, and evaluation to deploy custom AI at Meta Llama 3.1 NVIDIA NIM Microservices.

52

1 Comment

More TI (ServiceTitan AI initiative aka Titan Intelligence) features were also announced during ServiceTitan CEO, Ara Mahdessian, keynote including our first Agentic AI solution, The Virtual AI Agent, within Contact Center Pro, which captures calls on nights and weekends and decreases the call center burden by half. We also released to all Phones customers Second-Chance Leads, TI-powered feature that identifies calls that could have been booked by the call center but weren’t — and alerts contractors that they should call back to win the job. Ara called it “one of the most powerful capabilities we have ever built at ServiceTitan." https://lnkd.in/ginaepVK

14

Interesting that on the MIMIC IV readmission models, the LLMs get closer to the traditional ML models. Perhaps because LLMs are not great at representing numeric values (especially at the model sizes tested) that are informative of the prediction for acute outcomes, whereas readmission relies more on diagnoses and other more linguistically-coded inputs. (Though that doesn't explain the poor performance on MIMIC-III readmission)

3

🚀 Looking to streamline fine-tuning for models like Llama 3, Mistral, and Mixtral? Our latest blog shows how Anyscale simplifies LLMs fine-tuning with Ray: 🔑 Key takeaways: • Scalable fine-tuning with LLM-Forge • Model Deployment with RayLLM Read more: https://lnkd.in/gQwD4A2U

34

Meta's open-source Segment Anything Model 2 (SAM 2) is revolutionizing computer vision with its powerful image and video segmentation capabilities. Imagine isolating objects with pinpoint accuracy using just a single input point – that's the power of SAM 2. I recently witnessed this firsthand while segmenting a picture of Bella, our puppy. What's even more remarkable is that I've been running a version of SAM 2 on standard CPUs, despite the typical need for powerful GPUs. This opens exciting possibilities for bringing SAM 2's advanced features, including seamless object tracking in videos, to smaller devices like Raspberry Pis. Learn more about SAM 2 and its potential here: https://ai.meta.com/sam2/ #AI #ComputerVision #SAM2 #MetaAI #OpenSource

12

1 Comment

So much fun talking about the power of LLMs with Amazon Bedrock and Anthropic Claude. In summary, the ever-evolving landscape of GenAI, it's becoming clear that there won't be a single LLM (Large Language Model) to rule them all. The rise of application-specific LLMs heralds a new era, where various models will carve out niches to excel in. This diversity opens up a wide playing field for innovation and tailored solutions. When it comes to deploying AI models, striking the right balance in evaluation is paramount. We must weigh factors like cost, speed, and precision to ensure optimal performance and resource utilization. Finding this equilibrium ensures that our GenAI solutions are not just efficient but also effective in real-world scenarios. Anthropic Claude emerges as a frontrunner in the realm of LLMs, distinguished by its exceptional capabilities and groundbreaking advancements. Boasting unparalleled scale, efficiency, and adaptability, Anthropic Claude represents a pinnacle of GenAI innovation. Its ability to seamlessly integrate with Amazon Bedrock amplifies its prowess, providing users with an unparalleled ecosystem for knowledge integration and generation. With state-of-the-art architecture and a commitment to pushing the boundaries of what's possible, Anthropic Claude stands as a testament to the transformative potential of GenAI in shaping the future of technology and beyond. Prompt engineering holds the key to maximizing the potential of applications. Crafting prompts that consider persona, context, tags, and positioning can significantly enhance the model's ability to comprehend and generate relevant responses. This strategic approach empowers AI systems to better serve their intended purposes. In the realm of GenAI architectures, a robust knowledge base is essential. Amazon Bedrock stands out as a key component for architectures like RAG (Retrieval-Augmented Generation), providing a solid foundation to fuel intelligent decision-making and content generation. As we march towards AI integration, bridging the gap with legacy applications is crucial. Agents equipped with open APIs offer a seamless way to facilitate communication and interoperability, ensuring a smooth transition into the AI-driven future. Let's continue to explore, innovate, and collaborate as we navigate the exciting frontier of GenAI together! #aws #LanguageModels #Innovation #TechTrends #AIIntegration #webdevcon #claude #bedrock

68

There is so much research going into bringing the power of LLMs into the Enterprise context. The main techniques are k-shot learning or prompt engineering, RAG (Retrieval Augmented Generation) & Model Fine tuning (not to mention many variations of these e.g. Agentic RAG, GraphRAG etc.) However, I came across an interesting take on this from IBM and they are calling it "InstructLab" - where the claim is that they are able to teach the LLM "new skills" grounded in an Enterprise context. On the surface it looks like a k-shot learning + fine turning on steroids. Approach seems to be to utilize a small "skills training dataset" in order to generate a large synthetic dataset (perhaps utilizing a specialized LLM) that is then incorporated into the LLM weights natively. This could be a very interesting concept as long as the cost of adding new skills is not very high (when compared to fine tuning). Exciting times! #genAI #enterpriseAI

40

2 Comments

I wrote an article with David Yuan and Diego Z. from Tidemark, where we dove into how AI is shaking up the data game. For years, the big players—those with massive datasets—had an edge because their data acted like “gravity,” pulling customers in and keeping them there. But AI is changing that. Now, with tools to integrate and make sense of data across systems, startups can compete without needing to play by the big guys’ rules. It’s a new era of agility where data orchestration may just be the next big advantage. Exciting times for those rethinking their data strategy! https://lnkd.in/eqGDACYW #AI #DataStrategy #Innovation #Partnership

24

2 Comments

Dive into the architecture of OCI’s Generative AI Agent service, Retrieval-Augmented Generation (RAG) to enhance the contextual accuracy of AI agents, addressing the limitations of traditional LLMs and reducing hallucinations.

4

1 Comment

It has been my view for the last few months that the next big change in LLMs will be algorthmic optimization. Now comes this article. Don't sell your nvidia stock just yet, but remember that LxM technology is not just GPUs. https://lnkd.in/gyifxnHd

1

Watch this video to understand how LLMs store facts. The entire series is excellent to get an intuitive understanding. It all boils down to basic linear algebra. About 18 minutes into the video, it touches on sparse auto encoders and why LLMs generalize so well as the dimensionality increases

4

Exciting news for customers: Fine-tuning for Anthropic’s Claude 3 Haiku model is now available in Amazon Bedrock—currently the only managed service to offer this capability. This means our customers can now customize Claude 3 Haiku with their own data to build GenAI applications tailored to their specific needs, while keeping their data secure. Since we launched the preview a few months ago, we've already seen some great use cases. One of our customers, Robin AI, a legal AI company, is using this feature to optimize legal workflows and help lawyers speed up contract processes. With a fine-tuned Claude 3 Haiku in Bedrock, they are helping lawyers finish contract review processes 90% faster. Will be great to see how other customers use this new capability. https://lnkd.in/g2jU8Kfv

324

5 Comments

Getting evals right is the first step to building a working, production agent system. We help to automate the *building* of evals!

108

1 Comment

So what exactly do we do at Ringer Sciences? In short, we help companies build brand reputation intelligently. We do this by tapping the best human analysts and data scientists paired with strategic AI-enabled tools, technologies and techniques. We help uncover actionable intelligence in service of business results and architect next-generation digital experiences with data and insight serving as the foundation. Have a look at our breakdown of services and let's find a way to work together! CC Bob Pearson David Gallagher Ritesh Patel Jeremiah Owyang Yash Gad Tim Hayden

12

1 Comment

⛓️ Baseten Chains is a solution for reliable, high-performance inference for workflows using multiple models and processing steps. In other words: Chains is built for compound AI systems. 📈 We've been seeing more and more how compound AI is powering AI products. Check out Rachel Rapp's new post on our blog to learn more about what compound AI is, and why it's taking over the field: https://lnkd.in/ep22tupp

14

Responsible and accountable AI models are essential as generative AI continues to revolutionize industries. The rise of this technology, however, presents challenges in understanding how these models make decisions. Mechanistic interpretability addresses this by going beyond traditional methods that track statistical relationships. A recent research work by a team from #Anthropic has involved visualizing which neurons activate in response to specific prompts, advancing monosemanticity—ensuring each neuron’s activation correlates to a single, clear function. This step is crucial for fostering more transparent and safer AI systems in various applications. This is a great progress toward interpreting the thinking mechanisms of large language models (LLMs). #GenerativeAI #AI #LLMs #GenerativeAI #Grainger #MachineLearning #DeepLearning #ExplainableAI #DataScience

45

2 Comments