Mik C.

Anthropic just launched Claude 3.5 Sonnet, which beats both OpenAI GPT-4o and Anthropic’s previous paid model Claude 3 Opus on most benchmarks. Claude 3.5 Sonnet can execute and debug code, and has improved image understanding capabilities, including text transcription. A new feature called “Artifacts” enables users to view generated code, documents and websites in a separate window next to the chat. Claude 3.5 Sonnet is available for free to all users, but paid users get higher usage caps. Did Anthropic just take the lead back from OpenAI? It seems that ChatGPT’s Code Interpreter tool, which used to be one its main advantages, has finally met its match, but ChatGPT still has the advantages of real-time Internet access and custom GPTs which Claude lacks… What do you think? https://lnkd.in/d_hk86cf

10

3 Comments

AI-powered web agents fail. A lot. Why? Because they are inherently probabilistic tools operating in a big, complicated environment that changes all the time – the internet. At High Dimensional Research, our mission is to transform web agents from an unreliable, expensive pipedream for AI specialists into an efficient, production-ready tool for every full-stack developer. Today, we took a big step towards that goal by launching the Memory Index. This functions as a repository of user-anonymized web trajectories -- storing page structures as a graph, and navigation across those structures as searchable data. What does that mean? Well, basically, we observe the agents using our platform and transform their patterns of activity into a roadmap for the internet. The more people who use it, the more detailed and useful the map becomes. This way, an AI never has to figure out how to solve a problem that has already been solved. We just give them the answer. This radically reduces the failure rate, cost, and carbon footprint of running an AI agent. Check out our launch announcement and sign up at hdr.is/memory. https://lnkd.in/ePmCcScM

3

In the fast-paced world of Henerative AI, deploying Large Language Models (LLMs) into production presents a unique mix of challenges and opportunities. Irena Grabovitch-Zuyev and Suchita Venugopal from PagerDuty will share their journey of integrating LLMs into the product line within an ambitious two-month timeframe. 💡 We’ll Cover: ✅ Systematic Approach: Identifying LLM use cases, validating feasibility, engineering effective prompts, and developing a comprehensive testing strategy. ✅ LLM Service: A custom solution for secure and efficient LLM access, emphasizing robust security protocols, customer data protection, flexible LLM model switching, and redundancy for provider outages. ✅ Key Takeaways: Effective processes and strategies for rapid LLM deployment, highlighting non-negotiable aspects such as security measures and customer data protection. ✅ Lessons Learned: Honest discussion of the mistakes we encountered and the valuable lessons we learned to help you avoid similar pitfalls. This is a candid look at the challenges and triumphs of integrating generative AI at speed. #GenerativeAI #LLMs #ProductionDeployment #GenerativeAI #LLMDeployment

9

We are excited to share that we are back with our Summer School Series! Join us on August 8th for the Observability session of the series where we'll dive into advanced SLO management, enhanced synthetic monitoring, the Elastic AI Assistant for Observability, and universal profiling for .NET on Linux. Don’t miss out on the live demos, best practices, and a Q&A session where all your questions can be answered. Register now to secure your spot and elevate your Elastic Observability skills! https://gag.gl/tc7ncF #Elastic #ElasticSummerSchool2024 #ElasticObservability #observability #O11y #GenAI #Logs #AI #AIAssistant #SLOs #devops #techtalk #linux #monitoring #universalprofiling

1

"Productization is an affirmation," writes Adrian Bridgwater, "when any good or service evolves to a point where it can be productized and boxed, it denotes a certain solidity and branded consumability." In our conversation, I was delighted to share how #SAS equips organizations with readily deployable AI technology to productionize real-world use cases. https://lnkd.in/enUGCgAi #AIML #sasusers

9

Protecting users and developers on #GooglePlay is a top priority. In 2023 alone, we prevented 2.28M policy-violating apps from being published on Google Play in part thanks to our investment in new and improved security features, policy updates, and advanced machine learning and app review processes. Learn more about our initiatives that help keep you safe on Play and #Android: https://lnkd.in/g4-b2xYq

93

1 Comment

https://lnkd.in/d5R3Zfhn if you love MacOs as I do, you will be frustrated by lack of native support for GPUs (e.g. no cuda). things changed with MLX natively supporting M1, M2, and M3. The MLX examples repo has a variety of examples, including: Transformer language model training. Large-scale text generation with LLaMA and finetuning with LoRA. Generating images with Stable Diffusion. Speech recognition with OpenAI's Whisper.

17

I ran across this post in the Changelog newsletter: https://lnkd.in/eCdjVekq - I like the name slop for unwanted AI-generated content. I love LLMs. They are part of my daily workflow, and I've been exploring them more seriously while on my sabbatical. I wouldn't misrepresent generated content without critical review as my own, but the web seems to be filled with more generated garbage (or 'slop')—content that looks fine at first but isn't accurate or helpful. LLMs (and other generative AI) should help us to make things of higher quality, not just churn out slop more quickly.

11

1 Comment

What an event Google I/O has been. Excited to have new models launched in both AI studio and Vertex AI. Personally, the 1.5. pro flash is a must-try, and will- very-likely-stick model for developers and enterprises. The reasoning capabilities at par with some of the largest models in the market, but price and latency of a small model. What else could you ask? And Large context windows with multimodality are the cherries on top of the pie. If you have not already, I recommend you try it soon. Huge congrats to many Google teams for these launches. More below. https://lnkd.in/eYJMqUwS

167

4 Comments

As the volume of code continues to grow exponentially, software developers increasingly spend their time bogged down in the weeds of securing and scaling software. Today, we are announcing a new partnership with JFrog, so every developer can manage code and binaries more efficiently on two of the most widely used developer platforms in the world. Just as we did with Copilot Extensions, we are integrating the best of the software ecosystem within GitHub — to give the best possible experience to every developer on the planet. Build, secure, innovate, all from one dashboard. Let’s build from here, JFrog! 🐸

367

4 Comments

LLMs are designed to make stuff up. In most applications this an acceptable risk, after all, people make stuff up too. But in regulated, risk averse industries (finance, healthcare, insurance, legal) even a small mistake can cause big problems. Manual evaluation and monitoring is a crucial yet often overlooked component to catching errors and mitigating risk. Automated oversight can help, but there's no such thing as "Goodbye Hallucinations" unless humans are firmly in the loop.

10

1 Comment

Thrilled to add Gemini optional param mapping - https://lnkd.in/dyGCQQxU (+4 more updates 👇) 🛠️ Anthropic exception mapping - tool call error 💪 Add extra kwargs for text-compl calls 🪢 Fix - langfuse trace url on slack alerts 👀 PR - dynamic TPM allocation per project

9

Smaller and more informed (#RAG/LORA) models are faster and more useful in business practice. You shouldn't treat #LLM's as a general purpose silver bullet or you will have never create a product of any value. Always pick a composite of the right model and data for the problem. This three point strategy (Model, #Data, Problem) will help you create better results for you and your consumers. Don't be afraid to chain results together to generate a larger encompassing solution.

3

Here come the SLMs (small language models)! The ultimate architecture for a future app would be having local SLMs handle 90% of the work while calling one or more cloud LLMs for the rest. Ideally, the next step would be for operating systems or even browsers to come with a few default models optimized for the device so that we don't end up with a bunch of 5GB+ downloads. https://lnkd.in/gv6HaByu

5

2 Comments

Google Cloud and Anthropic made some exciting announcements: 📣 Claude 3 Opus, Anthropic’s most intelligent model, is now generally available on Vertex AI ⚙️ Tool use is available for the Claude 3 model family on Vertex AI, enabling Claude to act as an autonomous agent 📈Provisioned throughput is available for the Claude 3 model family on Vertex AI

19

1 Comment

Learn how to get up and running with Claude 3.5, the newest model from Anthropic on #VertexAI with Weights & Biases Weave! This tutorial is designed to help you get started with the Claude 3.5 Sonnet model on Google Cloud’s Vertex AI using Python. #aiml #generativeai #googlecloudpartners

1

I had a sneak preview of this "Cell-Based Architecture" InfoQ emag, and it's pure gold for architects! A big tip of the hat to Rafal Gancarz and the team 🙌 - How Cell-Based Architecture Enhances Modern Distributed Systems, by Erica Pisani and Rafal - Taking Advantage of Cell-Based Architectures to Build Resilient and Fault-Tolerant Systems, by Yury Niño Roa - Securing Cell-Based Architecture in Modern Applications, by Stefania Chaplin - Architecting for High Availability in the Cloud with Cellular Architecture, by Chris Price - Cell-Based Architecture Adoption Guidelines, by Guy Coleman https://lnkd.in/edDS5Um9

38

8 Comments

If you haven't watched the fireside chat between Patrick Collison, CEO of Stripe, and Jensen Huang, CEO of NVIDIA, I highly recommend taking time to watch it. The discussion delves into valuable leadership and management lessons shared by these two amazing CEO's & leaders. It's great to hear Jensen Huang's philosophy on managing direct reports, why he doesn't conduct 1:1s, utilizing feedback for learning and value of discussing feedback in a group setting, and not giving up on team (Jensen doesn't believe in fire fast approach). If you have been reading lot of management books for the last 30 years, lot of his advise at first will sound counter intuitive but as you internalize and hear his thinking behind it, you will appreciate his leadership style. Again one size doesn't fit all but great to listen to this interview. Watch the interview here: https://lnkd.in/gTnSU3NZ.

35

At the core of effective engineering leadership is the ability to stay connected with your team. This is perhaps the most challenging aspect because retaining the respect of your people is difficult unless you remain technically involved yourself. In my journey as an engineering leader, staying true to the individual contributor role that originally earned their respect has kept me close. This approach extends beyond technical expertise; it's about maintaining a deep connection with the core activities that propel product development forward at Gearflow. One of my biggest professional fears is becoming irrelevant or disconnected from my team and our work. This fear motivates me to continually reinvent and retool in response to market demands. Next month, I'll be sharing about my adventures generating synthetic data recently. And while I'm no expert when I start preparing for these talks, the process is incredibly valuable. It's less about delivering a perfect talk and more about how the preparation expands my thinking and exposes me to new capabilities. This helps me connect the dots to opportunities at Gearflow that bring real innovation to our customers. Speaking regularly holds me accountable and pushes me to learn ahead of the curve, ensuring I remain relevant and impactful day-to-day. Every talk is a journey of discovery, helping me to lead with humility, communicate clearly, and inspire a culture of innovation through a deep, technical understanding of our work.

19