Deep Learning and Machine Learning based explainable generative fashion design system for pre-production market optimization, demand prediction, customer preference analysis, and closed-loop design refinement.
AI-powered Loan Advisory Agent built with RAG to transform complex lending policies into clear, reliable answers on loan eligibility, EMIs, interest rates, fees, and more—grounded in policy documents with source citations.
AI-powered knowledge and productivity assistant with Smart Prompt Engineering, RAG-based PDF assistance, conversation memory, response evaluation, and personalized study tools.
Where feelings meet AI poetry. Generate original poems from opposing emotions and moon-phase writing rituals — built with Next.js and the Vercel AI SDK, crafting raw, sincere verses for whatever you're feeling 🌙 · ✒️ · 🌙
CodeGuru is a local AI-powered code teaching assistant built with Ollama, Python, and Gradio to solve coding problems, generate code, and explain programming concepts.
Journal Atelier — a Security-first Personal Gemini Journal — reflect & brainstorm with Gemini 3.6 Flash, multi-agent synthesis, mood trends, and an AI Project Studio, on Firebase Auth + Firestore + Cloud Run with the API key server-side only.
A secure, multi-tenant AI journaling app with contextual memory and semantic search — built on a custom security constitution defined in Google AI Studio, deployed on Cloud Run with Firebase Auth, Firestore, and the Gemini API. #AccelerateAIwithCloudRun
MCP server for hh.ru: search, resumes, applications, employer chat and hidden analytics over one applicant account, using the site's own JSON (no official API). Source-available, non-commercial.
Bengali-language flood risk assistant for Bangladesh using Gemini, Google ADK, and BigQuery window-function analytics on real GloFAS river discharge data.
An intelligent, agent-controllable edge-computing NVR powered by local AI. Features YOLO detection, CLAP audio triggers, Whisper transcription, InsightFace face recognition, Ollama vision, anomaly detection, Home Assistant MQTT, and a permissioned AI agent API.
Personal AI agent on Cloudflare Workers: Durable Object per conversation, markdown memory in R2, vector recall in SQLite, fan-out deep research. Free-tier hosting, measured limits.
The open, honest database of AI subscription plans — compare ChatGPT, Claude, Gemini & 80+ tools side by side, then optimize your stack to cancel what overlaps.
CLI for browser automation: let any AI agent drive your real Chrome — logged-in tabs, zero deps, no relaunch, no MCP server. A11y snapshots + element refs, CDP screenshots, network capture. 让任何 AI 智能体(Claude/Kimi/Qwen/DeepSeek/GLM)驱动你真实的 Chrome 浏览器——已登录会话、零依赖、无需重启。
Engineered to transcend human limits. While other platforms struggle with basic utility, we have perfected absolute dominance. Standardizing perfection since 2026.
A CPU-friendly AI chatbot built with Python and Hugging Face Transformers — includes knowledge-base search, a Gradio web interface, guardrails, and persistent memory.
Self-hosted AI chat bot where every answer carries a receipt: which instruction fired, what the API returned, which rows were read, and what it cost. Any provider — OpenAI, Anthropic, Ollama, vLLM.
Intelligent elder care health platform combining Scikit-Learn ML risk classification with Generative AI clinical summarization and FastAPI microservices.
Qwen3.8-27B on one laptop CPU: up to 2.52 token/s, 8 GB tested, no runtime accuracy loss. Native OpenAI-compatible API with function tools; no GPU or Python. | 单颗笔记本 CPU 运行 Qwen3.8-27B：最快 2.52 token/s，最低 8 GB 内存可运行，推理加速不牺牲准确性。原生 OpenAI 兼容接口支持函数工具；无需 GPU 或 Python。
The AI photo-compression meme, built for real: llmPEG turns images into ~1–4 KB semantic artifacts and regenerates scenes from text alone, with measured ratios reaching 802:1.
A RAG-based textbook tutor using document parsing, semantic chunking, embeddings, ChromaDB retrieval, and LLMs for context-grounded question answering.
An AI-powered Applicant Tracking System (ATS) built with Streamlit and Google Gemini Flash to scan resume PDFs against job descriptions, calculate match percentages, and detect keyword gaps.
ResearchMate is an agentic AI research assistant that uses multiple specialized AI agents — each responsible for a distinct part of the research workflow — instead of relying on a single LLM call to search, verify, summarize, and write.
62 rules that stop a generative step quietly replacing a brand's identity with something plausible and wrong. Every rule carries the incident that produced it.
Autonomous multi-agent AI research system using LangChain, Groq, Tavily, web scraping, and Streamlit to search, analyze, write, and critique research reports.
AI-powered interactive whiteboard that transforms natural language into fully editable Excalidraw diagrams. Built with Next.js, Clerk, Neon, and Drizzle
Navix AI is an intelligent browser agent that understands websites, automates web tasks, and helps users research, analyze, and interact with the web using AI.
An AI receptionist that lives on the business owner's Android phone — reads chat notifications, replies inline, books, takes orders, confirms payments.
Local, auditable coding agent for VS Code, optimized for small LLMs via llama.cpp with compact tools, batch patching, validation guards, metrics, handoff and forensic audit.
SarembokVE is an independent AI-native computing environment with its own runtime, agent fabric, persistent memory, system/OS layer, kernel technology, distributed compute architecture, and high-fidelity digital-human embodiment.
Melodic Software — Agent Plugins (agent-plugins.org) v1 packaging evaluation: parallel conversions of the org's Codex and Cursor plugin packages, the conformance validator, and the migration evidence ledger.
یک پکیج جامع لاراول برای راهاندازی چتبات هوشمند با پشتیبانی از درایورهای مختلف (AvalAI, GapGPT, OpenAI, Gemini, Local)، سیستم جستجوی دیتابیس (RAG بومی)، گاردریلهای امنیتی و تنظیمات داینامیک.
Cross-platform AI tool for grouping, organizing, and reporting product photos using Gemini, OpenAI, or Anthropic, with GUI, CLI, key rotation, and resumable processing.
