Оптичне розпізнавання символів Jobs

We are looking for an experienced backend development team to build a scalable AI-powered data aggregation and document intelligence platform. The system will automate data collection from multiple public sources, process documents using OCR and AI models, and provide intelligent search, analytics, and reporting capabilities. Responsibilities Develop scalable web crawling and data collection systems Handle dynamic websites, authentication flows, and anti-bot protections Build document processing pipelines for PDF, DOCX, XLSX, and scanned files Implement OCR and AI-based structured data extraction Create semantic search and analytics functionality Build REST APIs and backend microservices Develop notification and alert systems Implement monitoring, logging, and deployment automation Ensure...

I am ready to commission a full-scale Electronic Document Management and Archiving System for a leading transport operator in Saudi Arabia. The project begins with scanning and digitising roughly 10,000 – 50,000 legacy files—primarily administrative documents and financial records—and capturing them with Arabic/English OCR. Each file must carry the three core metadata tags we rely on: Date of creation, Department, and Document type. Once digitised, I need the EDMS configured for enterprise­-grade governance: role-based access, detailed audit trails, strict version control and immutable backups, all running in an HA topology that satisfies PDPL requirements. SSO and MFA must tie seamlessly into Microsoft 365, while automated workflows and data exchange connect back ...

We are looking for an experienced React Native developer to build a modern AI-powered Expense Management and Receipt Scanner mobile application. The app will allow users to scan receipts, automatically extract data using OCR, and track expenses with smart AI-based categorization and insights. This is an MVP project with potential for long-term collaboration and future feature expansion. Key Responsibilities: * Develop a clean and modern mobile application using React Native * Integrate OCR for receipt scanning and data extraction * Implement AI-based expense categorization and smart insights * Build dashboards with expense analytics and reports * Ensure smooth performance and user-friendly experience Core Features: * Receipt scanning via camera and gallery upload * OCR-based data ext...

create offical what app api from meta

I need a small, self-hosted OCR service written in C# (.NET 6 or later) that accepts a Base64-encoded image and instantly returns the recognised text. Tesseract must be the engine under the hood, configured for English only, with any preprocessing kept lightweight so the full request/response cycle stays in the low-millisecond range on a typical workstation. The DLL or minimal Web API you build has to be thread-safe so several parallel calls will not block each other. I’ll plug this straight into existing .NET applications, so please expose a clean public method or REST endpoint, document every parameter, and keep external dependencies to a minimum. Deliverables • Complete, well-commented source code and solution files • Build/run instructions and NuGet package li...

# REQUEST FOR PROPOSAL (RFP) # STAND-ALONE HOUSEHOLD GOODS MOVING COMPANY MANAGEMENT SOFTWARE ## Project Overview We are seeking a stand-alone desktop software solution specifically designed for a Household Goods Moving Company. The system must manage the entire shipment lifecycle from booking through final delivery, invoicing, collections, reporting, and profitability analysis. The software should be optimized for military, National Van Lines (NVL), NFC, COD, interstate, intrastate, and warehouse/storage shipments. The system must significantly reduce manual data entry by utilizing OCR and AI technology to scan trip sheets, bills of lading, and registration documents and automatically populate shipment records. --- # 1. CUSTOMER MANAGEMENT The software shall maintain a complete cu...

I have a collection of handwritten notes that were captured with my phone and saved as JPEG images. I need those notes turned into clean, editable text. The handwriting is generally clear, but I still want each line double-checked so the final document matches the original wording, spelling, and basic layout. I will upload the JPEGs to a shared drive and let you know the preferred order. You can use OCR as a starting point if you like, yet every page must be proof-read manually—accuracy matters more than speed. Deliverables • One neatly formatted Word or Google Docs file for each original image, preserving headings and bullet points where they exist. • A single consolidated file that combines all pages in sequence. • A short note highlighting any words you were u...

I have a batch of scanned PDF files that contain purely textual information. My goal is to have every word from those scans transferred into an editable digital format with 100 % accuracy, preserving original spelling, punctuation, and basic layout (paragraph breaks, headings, and any bullet points that appear in the source). The files are straightforward text—no tables or complex graphics—so this is largely a matter of careful transcription and light formatting. You are free to use OCR software if that speeds you up, but I still need you to proof-read line-by-line against the PDF because errors must be eliminated before delivery. Deliverables • A single Microsoft Word document (or Google Doc) for each PDF provided, saved with the same base filename. • All text ...

I have a collection of digital PDF documents that contain columns of pure numbers—no text, just figures that must be captured exactly as they appear. Your task is to transfer every numeric value from each PDF into a clean, structured file of my choosing (Excel or Google Sheets is fine; let me know which you prefer). Accuracy matters more than speed here: every digit has to match the source PDF, and the final sheet should mirror the original order so I can run analyses immediately. If you spot any illegible or inconsistent figures, flag them instead of guessing; I will clarify. Deliverable • A spreadsheet populated with all numeric data from every supplied PDF, double-checked for transcription errors and ready for basic calculations or pivot-table work. Once the sheet ...

I have a collection of digital PDF documents that contain columns of pure numbers—no text, just figures that must be captured exactly as they appear. Your task is to transfer every numeric value from each PDF into a clean, structured file of my choosing (Excel or Google Sheets is fine; let me know which you prefer). Accuracy matters more than speed here: every digit has to match the source PDF, and the final sheet should mirror the original order so I can run analyses immediately. If you spot any illegible or inconsistent figures, flag them instead of guessing; I will clarify. Deliverable • A spreadsheet populated with all numeric data from every supplied PDF, double-checked for transcription errors and ready for basic calculations or pivot-table work. Once the sheet ...

I have between 50 and 100 texts, emails, and screenshots that need to be turned into a clean, court-ready evidence binder for an upcoming Landlord and Tenant Board hearing. The task is strictly administrative: pull every message, place it in a timeline ordered by relevance (not simply by date), label each item as an Exhibit, and then compile everything into a single, searchable PDF. I work primarily in Microsoft Word, so I’d like the working file set up there first—complete with headers, page numbers, and clear exhibit tags—before you export the final version to PDF. Basic OCR or text-recognition is a plus, but not mandatory as long as the finished binder reads well and prints crisply. Deliverables: • A Word document containing the fully ordered timeline, each item...

I’m building an iOS app that turns written content into lifelike audio for Indian languages. Version 1 must: • Recognise text in PDFs, images and direct typing through a reliable OCR engine (Google ML Kit Vision, Tesseract, or an equivalent you can train for Indic scripts). • Speak that text aloud in Gurmukhi (Punjabi), Hindi and Urdu using a neural text-to-speech pipeline with male / female voice options, variable speed and uninterrupted background playback. Natural pronunciation is critical; speech should sound close to a native broadcaster, handle numerals and mixed-script phrases gracefully, and maintain at least 95 % OCR accuracy on standard printed pages. I prefer Swift / SwiftUI for the UI layer, with modular code so the core OCR-TTS logic can later power an And...

I have a PDF document that contains all the text I need transferred into an Excel spreadsheet. Your job is straightforward: open the PDF, copy the textual content exactly as it appears—maintaining headings, paragraphs, and any obvious section breaks—and paste it into the matching cells I will provide in an .xlsx template. Accuracy is essential; I do not want typographical errors, skipped lines, or misplaced text. If you prefer to use an OCR or extraction tool to speed things up, that is fine, but I still expect you to proof-read the result so the final sheet reads exactly like the source PDF. Once finished, please send back the completed Excel file and keep the structure identical to my template so I can import it directly into our system. Deliverable: • One fully popu...

I have a PDF filled entirely with numerical data that needs to live in Excel exactly as it appears now. Every column, row, header and subtotal in the source file must carry over without a single transcription error, so attention to detail is critical. You will be working from the original headers already shown in the PDF—please keep them untouched and in the same order. The finished spreadsheet should mirror the visual layout of the original document, so cell widths, column breaks and the overall table structure must match one-to-one. Deliverable • One Excel (.xlsx) file that reproduces the PDF’s table layout and numeric values with 100 % accuracy. I’m happy to answer any questions about the file before you start, and I’ll verify your work by cross-che...