Обробка природної мови Jobs

Use Case A: The Conversational WooCommerce ManagerInstead of relying solely on document drops, the agent acts as an on-demand, chat-based store manager. You or your staff can simply message the agent in natural language to execute catalog Input: You type a command into your Hermes chat interface or a connected Telegram bot (e.g., "Mark all nappies Size 4 nappies as Out of Stock" or "Update the retail price of Brand X toothpastes to $6.50"). The Reasoning: Minimax acts as the natural language parser. It reads your text, identifies the intent (Stock Update vs. Price Update), and extracts the exact product name and new value. The Execution: Max Hermes launches the headless browser, navigates to the WooCommerce wp-admin dashboard, searches for the specific product, adjus...

I need a production-ready chat agent that can handle frontline customer support in real time, generating dynamic answers instead of fixed scripts. The bot must: • Draw live information from my customer database, product catalog, support articles, and a separate set of internal training documents. • Create quizes, or downloadable PDFs in the users chosen language • Interpret a user’s intent, pull the correct data, and reply conversationally in clear English. • Maintain context across a session, hand off to a human seamlessly when confidence drops, and log the full conversation for analytics. You are free to build with Dialogflow, Rasa, LangChain, OpenAI, or a comparable NLP framework as long as you can deploy it to our existing platform (JavaScript - SEEDL) a...

I need a hate speech detection code implemented using BERT embeddings and Graph Attention Networks (GAT). The data source will be forum comments. The model must achieve high accuracy. Requirements: - Implement hate speech detection using GAT - Utilize BERT for word embeddings - Source data from forum comments - High accuracy in detection is essential Ideal Skills: - Experience with BERT and GAT - Strong background in NLP and machine learning - Proficient in handling and processing forum comment data - Ability to optimize models for high accuracy

Je souhaite faire développer une application mobile centrée sur l’intelligence artificielle, inspirée d’Instagram mais offrant une expérience nettement plus poussée. L’idée est de proposer un réseau social orienté photo-vidéo où l’IA amplifie la créativité des utilisateurs tout en ouvrant de nouvelles possibilités de monétisation et de suivi de performance. Fonctionnalités clés attendues : • Filtres avancés pour photos et vidéos : génération et ajustement en temps réel grâce à des modèles de vision par ordinateur (style-transfer, correction d’éclairage, effets 3D, etc.). • Opt...

Start your proposal with the word "MOSHI-SPEED". If you don't, your bid will be automatically deleted as a bot response. We are looking for a Senior ML/AI Engineer to fine-tune our voice assistant built on the Moshi / PersonaPlex streaming audio architecture. The goal is to extend the model to support Russian + Uzbek (Latin/Cyrillic) while maintaining zero-latency native switching. What is already done: All datasets (instruction/dialogue formats) are fully prepared, cleaned, and ready for training. No data collection or labeling is required. Scope of Work: Tokenizer Audit & Extension: Optimize the tokenizer for the Uzbek language (both scripts). SFT Fine-Tuning: Apply LoRA/QLoRA adjustments using the provided datasets. Crucially, updates must change weights without...

I need an AI-powered inbound voice application that automates the entire medicine-ordering workflow by phone. As soon as a customer calls, the assistant should pick up, greet the caller, and use natural-language (Hindi & English or mix) understanding to capture drug names and quantities without forcing the caller into rigid menus. After confirming the items, the agent must gather the customer’s name, phone number, and delivery address, repeat the details for verification, then write everything—time-stamped order, caller ID, full transcript, and audio recording—into a central database that my fulfilment team can query in real time. The telephony layer can run on Twilio, SIP, or any comparable cloud voice stack, as long as calls reach the AI with minimal latency. For t...

Develop an AI-powered system capable of accurately converting handwritten clinical notes into structured digital text. The solution should leverage advanced handwriting recognition models optimized for medical documentation and automatically generate comprehensive, clinically relevant discharge summaries based on the extracted information. The system should support medical terminology, maintain high transcription accuracy, and integrate seamlessly with hospital workflows and electronic medical record (EMR) systems.

Project Description: Project "OmniHub" Cross-Domain Multimodal AI Consulting Ecosystem with Self-Evolving State 1. Project Overview Project OmniHub is an enterprise-grade, multimodal AI consulting hub designed to deliver expert-level, specialized assistance across various domains including Medicine, Law, Agriculture, Marketing, and Education. Unlike standard chat applications, OmniHub features a fully persistent memory tier (tracking preferences, policies, and facts across sessions) and an autonomous self-learning mechanism. The system processes multimodal inputs (Voice, Images, and Video) and orchestrates specialized domain agents via a unified context framework. 2. Core Operational Pillars The platform adapts its operational logic depending on the chosen specialized vertical:...

**Project FRIDAY – Seeking AI Engineer to Help Design an AI Wearable Assistant MVP** **Project Overview** I'm the founder of **Project FRIDAY**, an early-stage AI wearable assistant focused on improving independence for blind and visually impaired people. Our goal is to build a hands-free AI assistant that combines computer vision, speech recognition, natural language understanding, and voice guidance to help users: * Detect and identify nearby objects * Read signs, labels, and documents aloud * Answer questions about their surroundings * Provide basic navigation assistance * Continue functioning with limited internet connectivity through an edge-first AI architecture We recently received feedback from a startup pitch encouraging us to narrow the MVP, so our current focus i...

I’m building an AI-powered service that can read batches of PDFs, Word files, and spreadsheets, then instantly surface the information I ask for. The core functions I need are: • Accurate data extraction from those three file types • Concise English-language summaries generated on demand • Reliable keyword and phrase identification that feeds a searchable index Everything will run in English for now, so no multilingual processing is required. I’m open to whichever stack you prefer—Python, Node, or another language—as long as modern NLP libraries (think PyPDF, docx, pandas, spaCy, transformers, LangChain, etc.) are used cleanly and the code is well-commented. Here’s how I picture the workflow: I upload a document or folder, the system pars...

I need a Retrieval-Augmented Generation chatbot that can instantly answer employees’ questions about our internal HR policy documents. The bot must handle both text and voice input/output so users can type or speak their queries and receive the appropriate response in English. The first release will live inside our existing mobile app, so everything—from authentication to UI widgets—has to be optimized for iOS and Android deployment. A lightweight in-app webview is fine for the chat surface as long as it feels native and supports microphone access. Here’s what I expect: • End-to-end pipeline that chunks, embeds, and indexes my policy PDFs, Word files, and the occasional slide deck, then serves grounded answers using a modern LLM. • Clear separation b...

Requirement Document Project Title: Handwriting & Pen Mark Removal from Scanned PDF/Image (Preserve Printed Content) Project Overview I am looking for an experienced Computer Vision / OCR / AI developer who can develop a solution to automatically remove all handwritten content and pen-based markings from scanned PDF documents while preserving all printed content exactly as it is. The solution should work on multilingual documents, especially: • English • Chinese (Simplified & Traditional) The final output should be a clean PDF with all printed text, tables, images and layout preserved. Functional Requirements • Remove handwritten text, notes, comments, initials, signatures, numbers and paragraphs. • Remove underlines (single/double/curved/zigzag). • Remo...

Saya sudah memiliki kerangka dasar chat-bot di WhatsApp, namun alur percakapannya belum lengkap. Fokus saya adalah membuat bot mampu memberikan dukungan pelanggan secara otomatis—terutama untuk: • Menangani keluhan pelanggan dengan nada sopan, solusi terarah, dan opsi eskalasi manual bila perlu. • Memberikan informasi produk yang selalu diperbarui, mencakup harga, ketersediaan, dan fitur utama. Saya membutuhkan bantuan untuk: 1. Menyusun skrip percakapan yang natural dalam Bahasa Indonesia, termasuk pesan pembuka, pertanyaan klarifikasi, dan tanggapan berbeda sesuai kategori masalah. 2. Mengonfigurasi alur decision tree atau flow builder di platform WhatsApp API (atau tool sejenis yang Anda sarankan) sehingga respon terkirim otomatis 24/7. 3. Menerapkan logika peng...

Hi I will share the details with the shortlisted candidates. Thanks

I’m refining an AI module that powers a heads-up, visually oriented interface. The core engine is Natural Language Processing, yet right now its intent detection and response generation drift off-target too often. I need your help to push raw and benchmarked accuracy noticeably higher without sacrificing latency. You’ll dive into the existing Python codebase (TensorFlow and a light PyTorch utility are already in play), audit the current model pipeline, then propose and implement improvements—be that better tokenisation, a more suitable transformer architecture, advanced data augmentation, or smarter post-processing. The interface overlays results on a visual HUD, so clean, deterministic outputs matter; hallucinations or fluffed confidence scores show up instantly to user...

Develop an AI-powered cross-platform (Android, iOS, and Web) Food Intelligence Platform focused on Indian foods that integrates a comprehensive Indian regional food database, personalized nutrition recommendations based on user demographics, medical conditions, lifestyle, family history, anthropometric data, dietary preferences, and health goals, AI-assisted interpretation of laboratory reports and health parameters, personalized meal planning, Indian recipe reformulation for healthier alternatives while maintaining taste and cultural relevance, prediction of nutrient retention during different cooking methods, functional food and nutraceutical recommendations, packaged food analysis through barcode scanning with ingredient and nutrition evaluation, food technology education modules, susta...

I want to replace the manual prospecting we do today with a fully automated, AI-driven lead generation and qualification engine aimed at builders and architects who buy our products. The core objective is simple: the system has to bring leads in on its own and instantly tell us which of those contacts are worth my sales team’s time. Here is what I expect when the project is complete: • An end-to-end pipeline that finds, captures and stores prospects from the channels we already use, then moves them straight into our CRM. • An AI agent that applies clear, data-backed rules to qualify each lead in real-time—no human triage—flagging the ones that match our ideal builder/architect profile. • A concise dashboard or daily report so my team can act on quali...

Facebook AI comment postong on various pages and profiles

# AI Medical Receptionist – Project Requirements ## Objective Build an AI voice receptionist for medical clinics that answers incoming calls 24/7, books appointments, answers common questions, and transfers urgent calls to staff when needed. ## Features ### 1. AI Phone Receptionist * Answer incoming calls automatically. * Greet callers professionally. * Speak natural English with a human-like voice. * Handle multiple calls if possible. ### 2. Appointment Booking * Ask for the patient's name and phone number. * Show available appointment slots. * Book, reschedule, or cancel appointments. * Send SMS and email confirmations. ### 3. Clinic FAQ The AI should answer questions such as: * Clinic hours * Address * Parking * Insurance accepted * Consultation fees * Doctor availa...