Sweep’s Post

prompter.vim Use vim as a tool for efficiently design, debug and save your LLMs prompts. Transform the Vim editor into an efficient prompt engineering environment, effectively replacing proprietary providers Large Language Models (LLMs) web playgrounds like: Azure OpenAI Service Playground or OpenAI Playground. Prompter.vim, from version 0.2, uses LiteLLM as a LLM provider abstraction layer! LiteLLM calls all LLM APIs using the OpenAI format: Bedrock, Azure, OpenAI, Cohere, Anthropic, Ollama, Sagemaker, HuggingFace, Replicate (100+ LLMs). So you can use prompter.vim with a vast list of different LLM providers! https://lnkd.in/dCHJRbsU

👉 *Disclaimer: this is a personal stretch project and a work in progress* I have been spending a lot of my personal time (evenings and weekends) building projects and doing my best to learn as much as I can about #AI. Not just because I am wanting to learn about it, but because my job and the market also demands it. As a result, I have been leveraging my existing skills and building out a #golang CLI (since my background is DevOps and Developer Tooling) to help me understand how we can interact with LLMs at an API and SDK level - > https://lnkd.in/eGgbPTgs So far the CLI has the following functionality: ✅ Ask a general question in the terminal ✅ Local LLM support using Ollama (llama3, phi3, mistral etc..) for offline use ✅ Generate images using DALLE based on a prompt ✅ Translate from one language to another Tech stack: 👉 Language: Go 👉 Framework(s): spf13/cobra, LangChainGo, Azure OpenAI Go SDK, Ollama 🤔 My question(s) to you: How are you learning about AI❓ How are you using AI❓ .. Drop a comment! If you want to contribute, open up a PR or submit an issue - the more the merrier. Alternatively, star the repo and follow along! #azure #azureopenai #aoai #golang #cli #community #opensource #developer #api #sdk #learning #llm #phi3 #ollama #langchain

Clickops as Code

A few weeks back, I put in a request for folks to build out interesting applications with Google PaLM APIs. The goal was to target different types of applications (web, mobile, utilities, etc) that can integrate PaLM API to address an interesting requirement. Akshat Jaimini has delivered the first of those applications with a code repository and associated blog post. Akshat has applied the PaLM API to an interesting area ... Github Actions. His solution is to do Log Analysis using Github Actions and PaLM API. Imagine ways in which you could apply this. When you commit a code to a branch, you kick off a range of tests that product logs that could contain errors. How about automating analysis of those logs with the PaLM API to summarize the issues, provide recommended solutions and more. Neat ... isn't it? Akshat has written a detailed blog post on the same: https://lnkd.in/dcJ_k5YU The source code is checked into the Google Cloud Gen AI Application Template repository, a set of application templates that help you get started quickly with Gen AI on Google Cloud. The solution is present in "palm-api-apps" (https://lnkd.in/dnSJZUUz) folder. If you would like to submit an application, do reach out. Check out the full list of applications ranging from web applications, Github Actions, Chrome Extensions, Slack Integration and more for Google Cloud Gen AI applications: https://lnkd.in/de5WxrAu #GoogleCloud #GenAI #CommunityContributions #Community Prashanth Subrahmanyam Abirami Sukumaran Lavi Nigam Thu Ya K. Ankur Kotwal Akshat Jaimini

GitHub Copilot, Amazon Code Whisperer emit people's API keys ✅ GitHub Copilot and Amazon CodeWhisperer can be prompted to emit hardcoded credentials that these AI models captured during training. ✅ The researchers built a tool called the Hardcoded Credential Revealer (HCR) to look for such things as API Keys, Access Tokens, OAuth IDs, and the like. ✅ GitHub and Amazon did not immediately respond to requests for comment. More in Source: https://lnkd.in/eBV2KNF7 Bite-sized news made with ❤️ by https://savelikeapro.ai

Imagine if you could speak to your code like a person... 🤔 🗣️💻 Now you can! In a new blog post, we build an app that lets you chat with a GitHub repository. Explore technologies like FastAPI, LlamaIndex, and OpenAI! Topics include: RestAPIs, tasks queues, and LLMs. 🚀📚 Read the blog: https://buff.ly/3VEptme Source code: https://buff.ly/3PGStG3

Integration of Next.js with AI

Here is a template where langchain is completely integrated with next.js 13 . It includes features like chatbot , agents , talking to a PDF file ,getting a structures output etc. https://lnkd.in/dwCZDwiu #genai

Duet AI in GCP is now in preview. There is a 3 steps process to kick start... To Set up Duet AI for a GCP project: https://lnkd.in/grHXhJZd Being a Database Administartor, I am interested to kick start Gen AI - Deut AI related to databases in the following areas: Comments, suggestions, tips are most welcome!!! Excerpts from the link : https://lnkd.in/grik8T38 Simplify the development journey across your databases Duet AI helps you complete your data management tasks faster. With Duet AI in Cloud Spanner, you can get code suggestions to structure, modify, or query your data—generated from natural language. With Duet AI in Database Migration Service, we’re making it easier to modernize your Oracle databases to AlloyDB, especially for the last mile of code conversion. (Features currently in Preview.) Excerpts from the Link: https://lnkd.in/gkJ_mDni Accelerate and modernize databases Duet AI in databases helps you take developer productivity to new levels. We’re introducing this capability for Cloud Spanner today, and it will also be coming to AlloyDB and Cloud SQL. With Duet AI in Cloud Spanner, you can generate code to structure, modify, or query your data using natural language. For instance, with a simple command such as “write a query to show all data in the messages table,” you can have Duet AI automatically generate the required SQL for you to further review or modify, which can significantly reduce the time to develop code for applications.  In addition, to make it easier to modernize your databases, we’re bringing the power of Duet AI to our Database Migration Service (DMS). Duet AI in DMS helps automate the conversion of database code that could not be converted with traditional translation technologies, such as stored procedures, functions, triggers, packages, and custom PL/SQL code. You’ll be able to use Duet AI in DMS to migrate the last mile of Oracle to AlloyDB and Oracle to Cloud SQL-for-PostgreSQL later this year. Kudos to the author of the following link - alerternate to GitHub Copilot X - VS Code with Google Duet AI.

Build ChatGPT 2.0 with Next.js and OpenAI API - Chat Rooms, MongoDB, Dynamic Routes, Custom API