a
ahmad_mahad_

Ahmad Mahad

@ahmad_mahad_

RAG Chatbot Developer and AI Application Developer

Pakistan
Englisch, Urdu
Einige Informationen werden in englischer Sprache angezeigt.
Über mich
I build AI-powered web applications that turn business data into intelligent tools — specializing in RAG systems, LLM integration, and custom chatbots using Python, LangChain, and OpenAI APIs. I also develop full-stack web apps with the MERN stack (React, Node.js, MongoDB) and scalable Flask backends. My focus: clean architecture, fast delivery, and AI solutions that actually solve business problems not just demos. Let's build something your users will rely on daily.... Mehr lesen

Kompetenzen

a
ahmad_mahad_
Ahmad Mahad
offline • 
Durchschnittliche Antwortzeit: 1 Stunde

Meine Dienstleistungen

Regelbasierte Chatbots
I will build a custom rag chatbot using your business data
Ab50 $ / ProjektWeitere Details

Arbeitserfahrung

Self_Employed

Self Employed

Vollzeit • 3 mos

Intern Management Portal-Lab Project

Jun 2026 - Present • 1 mo

Designed and developed a full-stack intern management portal using the MERN stack (MongoDB, Express.js, React.js, Node.js) for a research lab. Built features for intern onboarding, task tracking, and record management. Deployed the application on AWS EC2, handling the project end-to-end — from database schema design to deployment and hosting.

Social Media Discourse Analysis-Research Work

May 2026 - Present • 2 mos

Led an NLP research project analyzing 14,000+ social media posts (Instagram and X/Twitter) to study public discourse patterns. Applied RoBERTa for sentiment analysis, LDA for topic modeling, and SentenceTransformer embeddings for semantic clustering. Produced a full analytical report and stakeholder presentation, translating model outputs into clear, actionable insights — directly applicable to building AI tools that understand real-world text data.

Self_Level

Local RAG-Based Study Assistant

Self Level • Selbstständig

Mar 2026 - Present • 4 mos

Built a privacy-focused RAG system that lets students query their own study material (notes, textbooks, slides) using a locally-run LLM, no data sent to external APIs. Designed the full retrieval pipeline: document chunking, embedding generation, vector storage, and context-aware response generation, served through a Flask backend API. Focused on running models locally for data privacy and zero ongoing API cost, making it practical for real student use rather than just a demo.