Developer builds AI automation system to manage MyZubster open-source project

·2 views

A developer has built an AI-powered automation system for MyZubster, an open-source platform focused on plant mapping, privacy-first payments, and human-centered AI. The system automatically monitors GitHub repositories, triages new issues and pull requests, and creates bounties for labeled issues without manual intervention. It uses locally running AI models — Gemma 2B, Llama 3.2, and DeepSeek R1 via Ollama — with automatic fallback between models if one fails. A Telegram bot integration provides real-time notifications and interactive commands for contributors, addressing the project's challenges with slow communication and manual monitoring. The modular system is built on a Node.js stack with MongoDB, Express, and React, and is designed to reduce the operational burden of managing a growing open-source community.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Developer migrates Next.js app from Node.js SQLite to Cloudflare D1 for remote access

A developer migrated a Next.js application from Node.js's built-in node:sqlite module to Cloudflare D1 in order to deploy it on Cloudflare Workers and access it remotely. The migration was necessary because the node:sqlite module is unavailable in the Cloudflare Workers runtime. The core work involved moving the database schema into a dedicated migration file and rewriting the database module to use D1's asynchronous API instead of node:sqlite's synchronous one. Parameter binding and row-change metadata required minor API adjustments, but the underlying SQL queries remained unchanged. The D1 database binding was accessed via getCloudflareContext() from the @opennextjs/cloudflare package, and all call sites were updated to use async/await.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Task Completion Protocol for AI Coding Agents Prevents Silent Failures

AI coding agents often declare a task 'done' after running only limited checks, leaving lint errors or failing tests undetected in the broader codebase. A software developer proposes a structured approach called the Task Completion Protocol, stored in a repository's AGENTS.md file, to enforce stricter verification before completion is claimed. The protocol requires agents to run the project's actual lint and test commands, fix any failures, and report concrete evidence of results. It distinguishes between coding and non-coding tasks to avoid unnecessary test runs, such as skipping backend tests for documentation-only changes. The goal is to catch broken states earlier in the development workflow, before they propagate to pull requests or subsequent agent sessions.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

TCS Prime Candidate Steers Interview Away From DSA to ML Projects

A software engineering candidate recently shared their experience navigating a 45-50 minute TCS Prime interview on the DEV Community platform. Facing the prospect of tough Data Structures and Algorithms questions, the candidate deliberately steered the conversation toward their own machine learning and full-stack development projects. The strategy involved proactively discussing project details, including JWT-based security and real-time Socket.io features, to keep the panel engaged on familiar ground. The candidate also candidly addressed a low starting CGPA and unconventional hobbies like anime and manga, which the panel received positively. The interview ultimately felt conversational rather than adversarial, and the candidate is now awaiting final results.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Spring AI Framework Lets Java Developers Add AI Features to Spring Boot Apps

Spring AI is a framework within the Spring ecosystem that enables Java developers to integrate generative AI capabilities directly into Spring Boot applications. It provides portable, provider-agnostic abstractions for features such as chat models, embeddings, vector stores, tool calling, and Retrieval-Augmented Generation (RAG). Developers familiar with Spring Boot can use components like ChatClient to send prompts to AI models without writing provider-specific code throughout their applications. The framework supports multiple AI providers, including OpenAI, through dedicated starter dependencies. Spring AI is aimed particularly at enterprise Java applications where maintainability and consistency with existing Spring architecture are priorities.

0 comments Read more at DEV Community