H2O.ai’s Post

Quickly collect user #feedback with #ratings. Rating controls allow you to collect structured, numerical data from users that can be easily compare and analyzed. Compare leading solutions for #Blazor.

Thanks to Chris O'Brien for an amazing write-up in Sifted on how PhotoRoom became Frances' hot young gen AI thing. 3 of the "magic ingredients" to PhotoRoom's stratospheric growth: 1. Relentlessly focus on customer feedback: If you're not getting banned from your local McDonalds, you're not doing it right. 2. Build for mobile rather than desktop: Mobile software is 10 times bigger than anything you’ll do on desktop. 3. Focus on speed: Speed matters to users on mobile, who expect results in milliseconds. Link to article in the comments

Considering the maze of bespoke systems, mobile, and web development choices, how do you validate, size, and make decisive steps forward? Enter the "Accelerated Business Case" workshop by FXBITS in collaboration with Pangea.ai Over two transformational days, refine your concept and sculpt your roadmap to progress with clarity and confidence. During this intensive hybrid session, you will: 🚀 Decode the intricacies of high-level system architecture 🚀 Chart a vivid customer journey map to visualize engagement 🚀 Acquire design guidelines complemented by illustrative sketches 🚀 Fine-tune a prioritized feature backlog tailored to your vision 🚀 Envision a robust potential delivery roadmap Dive deep, strategize, and transform fleeting ideas into concrete action: 🔗 https://lnkd.in/gnFdZ5r3 Join FXBITS and harness the potential of your business vision! #BespokeSystems #Prototyping #POCs #MVPDevelopment #TechConsulting #MobileDevelopment #WebDevelopment

It's fascinating to witness how advancements in one technology can be quickly implemented for creating efficiencies and new capabilities in another tech area. In this case, LLMs have been adapted to create dynamic user interfaces. Check out this video to learn more: https://lnkd.in/e_a7kCy4

Pretty cool paper by Tamer Abuelsaad, Deepak Akkil Ravi Kokku and team on their implementation of Agent-E [1,2] - a multiagent workflow (built with #autogen) for accomplishing web based tasks (driving web interfaces). Paper: https://lnkd.in/gNk-nUAY The paper includes lots of nuggets that help us improve on our collective understanding of how to build interface agents[3], including: - Hierarchical agent architecture with separation of roles e.g, planning, web navigation, etc and the benefits it provides (task verification, error recovery etc) - DOM distillation to reduce noise in web pages. - Results (based on evaluation on the WebVoyaGer benchmark) indicate that a text only representation the use (DOM) does outperform existing multimodal approaches. Great example of where a multi-agent but less powerful setup can result in overall better performance. References [1] Paper on arxiv https://lnkd.in/gNk-nUAY [2] Paper github repo: https://lnkd.in/gaXyuTje [3] Interface Agents - Building Multi-Agent Applications that Act via Controlling Interfaces (Browsers, Apps) https://lnkd.in/eVdFqc_J #autogen #multiagent #multiagentsystems

In essence Rivet is a no code to low code prompt chaining GUI which is focussed on building complex flows, which can split (run processes in parallel) and converge again. Data handling and transformation plays a big part in the Rivet UI, together with vector stores and KNN. Together with conditions, and numerous conditional nodes. Rivet refers to flows as graphs, as opposed to a chain, application or flow. Rivet does not have a vast array of choices in terms of vector stores, LLMs and the like. It does seem like Rivet’s focus is more on building complex flows, which can be optimised and re-used as components. Read more here: https://lnkd.in/eu_s-vzm #LargeLanguageModels #PromptEngineering #ConversationalAI

New in Predibase! ⚡ Fine-tuning is now up to 5x faster! 💯 All users can fine-tune with A100s at no extra cost 🛠️ View and manage deployments in the UI Read more on our blog: https://pbase.ai/3IZGWhl

ModelScope-Agent 🤖 is a new paper with code that introduces a new framework for working with agents and APIs. the framework offers a general and customizable way for connecting any LLm agent to any tool on different tasks. Their goal is to create an central hub for agents and tools. They support a variety of tools and APIs right of the box 🎁 Looks super easy to use and can be integrated with Langchain as well! worth checking out, enjoy