Amna Tiwana - codesphere

codesphere

Sign in Subscribe

Amna Tiwana

Product Marketing Engineer from Pakistan.

Smart Chat RAG: Building Document-Aware Conversational AI

Smart Chat RAG: Building Document-Aware Conversational AI

A modular chat system built on RAG principles enables transparent, document-grounded conversations. It uses semantic search, LLM inference, and preprocessing like OCR and table parsing for accurate responses.

Smarter PDF Comparison: How to Catch Real Changes, Not Formatting Noise

Smarter PDF Comparison: How to Catch Real Changes, Not Formatting Noise

comparison of PDFs by extracting content, detecting key changes, and producing a highlighted PDF and a detailed report.

Efficient Document Comparison for Compliance Using AI and Embeddings

Efficient Document Comparison for Compliance Using AI and Embeddings

AI-powered system for comparing internal documents with regulations using embeddings, chunking, and local LLMs. It avoids keyword matching pitfalls by understanding meaning, storing vectors in PostgreSQL with pgvector, and generating contextual insights

Model Context Protocol: Beyond Fragile Prompts to Secure AI Integration

Model Context Protocol: Beyond Fragile Prompts to Secure AI Integration

The Model Context Protocol (MCP) enhances AI workflows, ensuring reliability, security, and consistency in your projects. MCP helps you build agents and complex workflows on top of LLMs.

AI Summarization of Long Documents: Tackling Long Documents with Precision

AI Summarization of Long Documents: Tackling Long Documents with Precision

Long texts are difficult to summarize, but recursion can divide them into small parts. The approach is precise and preserves the meaning at any step of iteration.

Structure PDF Table Data for AI Applications with GMFT

Structure PDF Table Data for AI Applications with GMFT

GMFT is a fast, lightweight toolkit for extracting tables from PDFs into formats like CSV, JSON, and Pandas DataFrames. Leveraging Microsoft's Table Transformer, GMFT efficiently processes both text and image tables, ensuring high performance for reliable data extraction.

Extracting Metadata from PDFs with Named Entity Recognition using Spacy

Extracting Metadata from PDFs with Named Entity Recognition using Spacy

NER identifies entities like people and locations in text. SpaCy automates this with pre-trained models, offering accuracy, speed, and multi-language support. It excels at handling large datasets efficiently compared to rule-based methods.