Dave R.’s Post

Worth reposting here. 😅 https://lnkd.in/g4Kktqn6 I’m personally really excited for the soon approaching back half of the hype cycle where we can spend time adding some actual software engineering and scalability expertise to the decades of poorly written Python code from the ML world and start implementing features users want without slapping the widely abused phrase “AI” on everything. 🙌🏻 If you think it needs a chat bot, it doesn’t. If you think you’ll get more engagement or find market fit from a “✨ Try it with AI” button slapped on everything, it doesn't and you won't. The best AI-driven features are specific, solve real problems and are so seamless in your product that users don’t need a desperate “LOOK WE HAVE AI TOO” billboard to find them.

Enjoying jumping into generative AI and #LLM development. The GitHub repo below provides a hands-on example and walkthrough to quickly spin up your own #LangChain, GPT4 environment with a simple frontend in codespaces ...and (most) importantly, observing LLM application performance with code-level traces for LLM applications with #Dynatrace as you test and productionise. Link to repo: https://lnkd.in/g5W_WFdw

🚀 *Exciting News!* 🚀 Introducing *LOCAL JARVIS* 🤖 Unlock chatgpt's pro functionality without any limitations with *LOCAL JARVIS*! The kicker? You can do it all offline! This dream is now a reality. Utilizing the power of the Open source LLM, Vision models and Python's Django/Flask or Streamlit framework, I specialize in creating custom-built smart AI applications tailored to your business needs. Say goodbye to costly API charges and reduce expenses on AI application frameworks. Let's revolutionize your AI experience together! 💡 #AI #Innovation #Technology

🎉 After months of work, ~40k lines of code and a hefty bill for OpenAI API usage, I'm really thrilled to present Devra, an AI software engineer. Devra is a desktop app that works with any code you have, or can start a project with a prompt (in the link there's a video for "Create a JavaScript implementation of the game of Snake") Please check it out; it is free to use and I would appreciate any thoughts, suggestions, etc. I hope you love it! https://devra.ai

Cognition AI presents Devin, the first AI software engineer. Devin is capable of building apps end to end, finding bugs in production codebases, and even fine-tuning some AI models. Devin is the new state-of-the-art on the SWE-Bench coding benchmark, has successfully passed practical engineering interviews from leading AI companies, and has even completed real jobs on Upwork. Devin is an autonomous agent that solves engineering tasks through the use of its own shell, code editor, and web browser. When evaluated on the SWE-Bench benchmark, which asks an AI to resolve GitHub issues found in real-world open-source projects, Devin correctly resolves 13.86% of the issues unassisted, far exceeding the previous state-of-the-art model performance of 1.96% unassisted and 4.80% assisted. The future is here. Congrats to the Cognition team Scott Wu Walden Yan and Neal Wu for building an incredible product.🚀

Say goodbye to those frustrating hours spent hunting down small bugs in your code. Introducing AI for Troubleshooting & Resolving CI/CD Failures 🔍 Our bot can now identify and suggest solutions for pipeline review errors, making debugging faster and easier. Check it out 👉 https://hubs.la/Q02BngR00 #AI #DeveloperExperience #GitHub #CodeQuality #DeveloperTools

There are lots of interesting AI coding tool attempts happening in the space between AI chat, normal local IDE development (even with Copilot or Cursor) and into all of the usually really boring and hard back-end production stuff: @lovable_dev, @v0, @replit, @stackblitz and now Google IDX. I think the main question I have about these tools at the minute is: are they going to save me time and get it right over complex production coding projects or just make a mess and cause me to waste time fixing it all?

🚀 Excited to share my latest blog post on creating a chatbot with a .llamafile and integrating it with a Flask web frontend! 🌐🤖 In this post, I dive into the step-by-step process of setting up the model, creating a Flask API, designing a user-friendly webpage, and running the chatbot smoothly. Whether you're a beginner or looking to enhance your skills, this guide is for you! Check it out and let me know your thoughts! 📚💡 #Chatbot #DataScience #Flask #WebDevelopment #AI #TechTutorial

A very interesting point of view regarding different AI techniques for Devs and how to combine more traditional approaches with generative ai bearing in mind their respective strengths and risks. https://lnkd.in/eniXnY-v

Why context length is the key to better code generation 👇 Most LLM solutions optimize for chat: - Short prompts - Long responses - Limited context But coding is different... Code generation needs to understand: - Dependencies - Call sites - READMEs - Build files - Third-party libraries - Project context Our research shows: code prediction quality keeps improving well beyond 10k tokens of context. The challenge? Processing all that context while maintaining low latency. The numbers: - Chat: ~100 input tokens → 100s output tokens - Code: 1000s input tokens → 10s output tokens Different workload = different optimization needs Here's how we do it at Augment: https://lnkd.in/gVT5kQRg #AIforDevelopers #SoftwareEngineering #CodeGeneration #AI #DeveloperTools #Programming