Developer Builds AI-Powered Duplicate Detector for Appwrite Projects

·5 views

A developer has created Appwrite AI Duplicates Detector (AADD), a full-stack web application designed to identify and manage duplicate files and documents within Appwrite projects. The tool uses AI algorithms and perceptual hashing to detect duplicates across both Appwrite databases and storage buckets, addressing a gap not covered by Appwrite's native features. Built with Next.js, Flask, and the Google Gemini API, AADD offers a three-stage workflow covering project connection, duplicate scanning, and bulk management of results. The application also includes a gamified feature called AI Garden, which lets users track their data-cleaning progress through an interactive interface. A live version of the app is publicly accessible via Appwrite's hosting network.

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)

Same AI Model Can Cost 14x More Depending on the Provider Serving It

A new analysis of 405 paid models on the OpenRouter routing platform found that 182 models are served by two or more providers, with prices varying by a median of 1.87x and as much as 14.47x for the same model. Beyond cost, providers differ on numerical precision — fp4 versus bf16 — and uptime, factors that are not reflected in the model name or API call. Data residency is another hidden variable, as 39% of multi-provider models are served from locations in more than one jurisdiction, which matters for compliance obligations. Researchers also benchmarked latency across 329 models and graded 330 for Korean-language quality, finding that performance does not reliably track release date or parameter count. The findings, published on Hugging Face, argue that selecting a model and selecting an endpoint are two separate decisions, each with meaningful cost and quality implications.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Developer Builds 'Mandatum' to Make AI Agent Authority Traceable and Revocable

A developer has released an open-source library called Mandatum that addresses a core security gap in AI agent systems: the inability to trace actions back to a specific human authorizer. Currently, most AI agents inherit shared credentials like API keys or service accounts, making it difficult to identify responsible parties or revoke access selectively — a problem confirmed by two 2026 industry surveys showing the majority of organizations cannot distinguish AI activity from human activity. Mandatum creates a signed, hash-linked delegation chain rooted in a named human, so authority can only narrow as it passes down to sub-agents and revoking any link automatically cuts off everything below it. The library also enables sequence-aware authorization, meaning it can deny actions based on the order in which they occur — for example, blocking a write operation that follows a read from untrusted external content. Rather than making authorization decisions itself, Mandatum delegates that responsibility to existing policy engines like OPA or Cedar via the OpenID AuthZEN standard.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

React Re-renders: Why Memoization Is Often Not the First Fix

A common React developer reflex is to add React.memo, useMemo, and useCallback whenever a component re-renders too often, but this approach treats symptoms rather than root causes. Re-renders do not always mean DOM updates; React only modifies DOM nodes that actually changed during the commit phase, making many re-renders harmless. Performance only becomes a real concern when repeated rendering involves expensive computation or visibly slows down user interactions. A more effective first step is restructuring components so that local state lives as close as possible to where it is used, preventing unrelated parts of the tree from re-rendering on every state change. Memoization and related hooks remain valid tools, but should be applied selectively after identifying a specific, measurable source of wasted work.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Open Skill Library Aims to Fix LLM Blind Spots in Embedded Hardware Dev

AI coding assistants frequently generate firmware that compiles and flashes correctly but fails in practice due to board-specific hardware quirks they are unaware of. Common pitfalls include undocumented pixel offsets on cropped displays, shared SPI buses causing conflicts, and GPIO pins that exist in code but have no physical header connection. A developer has begun building an open collection of board-specific context files, called skills, designed to be loaded automatically by Claude Code when working with a matched microcontroller board. Each skill bundles a pin map, known hardware gotchas, in-depth peripheral references, and a verified working project template — all sourced from datasheets, vendor wikis, and real debugging sessions. The project targets the core issue that LLMs lack access to scattered, board-level knowledge that only surfaces through hands-on hardware experience.

0 comments Read more at DEV Community