Jahangir Abbas
BS Information Technology student at the University of the Punjab with a strong interest in software engineering, artificial intelligence, and modern web technologies.
Summary
Social Profiles
GitHub Activity
Engineering Mindset & Approach
Decomposing large systems into workers, queues, chunking strategies, and vector indices before writing code.
Focusing on production-ready products (Chaibook LLM, Equilibrium AI, NetLens AI) with guardrails and streaming.
Building asynchronous ingestion pipelines with S3, BullMQ, Redis/Valkey (Aiven), and Qdrant.
Creating responsive, Apple-inspired interfaces using React 19, Next.js 16, and Tailwind CSS v4.
Stack
Experience
Independent Consultant
AI Systems Architect
- Architected Chaibook LLM, a flagship document intelligence platform handling multi-format ingest (PDF, URL, YouTube) with Qdrant vector retrieval.
- Designed Equilibrium AI, a multi-LLM consensus & synthesis engine processing parallel model streams.
- Engineered Persona AI chat application with real-time persona switching and session memory storage.
- Built asynchronous background ingestion worker pipelines using S3, BullMQ, Valkey (Aiven), and Qdrant vector database.
- Delivered production web solutions for clients including automotive platforms, workshop management systems, educational tools, and AI SaaS applications.
Full-Stack & AI Systems Developer
Education
- • Currently pursuing undergraduate studies focusing on software development, programming, databases, networking, and modern computing technologies.
- • Actively studying software architecture, system design, data structures, algorithms, and emerging AI technologies to strengthen both theoretical and practical foundations.
Featured Projects & AI Systems
Chaibook LLM
AI & RAG SystemsNotebookLM-Inspired Document Intelligence & RAG Platform
Flagship AI application enabling users to upload PDFs, text documents, URLs, and YouTube videos for deep conversational context retrieval, source citations, and right-sidebar evidence verification.
Equilibrium AI
AI & RAG SystemsMulti-LLM Consensus & Synthesis Engine
An innovative multi-model AI system where one user prompt is processed simultaneously by three different LLMs, followed by a fourth meta-evaluation model that synthesizes their responses into one optimal answer.
RAG Subtitle Search
AI & RAG SystemsVideo Subtitle Indexing & Exact Timestamp Navigation
Upload video subtitles or YouTube transcripts to perform semantic context queries and jump directly to exact video timestamps with verbatim evidence highlights.
Persona AI
AI & RAG SystemsAI-Powered Persona Chat Simulation
An interactive AI-powered chat application that simulates conversations with popular Indian software educators Hitesh Choudhary and Piyush Garg. Supports real-time persona-based switching and session memory.
AI & System Architecture
Multi-Format Ingestion
Accepts PDFs, text files, URLs, and YouTube video transcripts via presigned S3 upload URLs.
{
"fileId": "doc_92871",
"fileName": "handbook_v2.pdf",
"fileSize": "4.2 MB",
"uploadUrl": "s3://jeem-ingest-bucket/uploads/handbook_v2.pdf",
"mimeType": "application/pdf"
}Qdrant Database
Dense + Sparse BM25 hybrid search with payload filters.
BullMQ & Valkey
Background ingestion workers deployed on Aiven Redis/Valkey.
Input/Output Gate
Cohere re-ranking and toxicity hallucination checks.
Certifications(6)
OpenAI API Deep Dive
OpenAI / Independent·2025
Next.js Production Patterns
Vercel Ecosystem·2025
Retrieval-Augmented Generation Systems
Jeem Labs — Internal Curriculum·2025
Advanced TypeScript for Application Architecture
Self-paced / Applied Practice·2024