
Hi, I am Abdul Azeem Hashmi, an AI & Backend Engineer studying BS Artificial Intelligence at FAST NUCES, Islamabad. I build autonomous AI agents, LLM-powered applications, and full-stack web experiences focused on performance and modern design.
Combining cutting-edge artificial intelligence with scalable, modern web application architecture.
Architecting autonomous agents, multi-tool reasoning workflows, LLM pipelines, and goal-driven AI workstations powered by Gemini, OpenAI, and LangChain.
Expertise in Gemini API, LangChain, Tool Calling & Agentic RAG.
Building production-grade, reactive web apps and scalable asynchronous backend APIs using Next.js, TypeScript, FastAPI, Flask, and Tailwind CSS.
Proficient in Next.js App Router, TypeScript, Python & RESTful APIs.
Designing robust relational and document database schemas with seamless persistence, real-time logging, and cloud data integrations.
Experienced with Supabase, SQLite, MongoDB, MySQL, and Docker.
Industry and academic research internships focusing on autonomous systems and full-stack software.
UAV Dependability Research Lab
Researched and engineered AI logic models for autonomous UAV systems, conducting dependability evaluations, waypoint route optimization, and rule compliance testing.
CODOC PVT LTD
Architected and implemented scalable software solutions, optimized full-stack application performance, and collaborated on modern agile software engineering workflows.
The tools, frameworks, and programming languages I leverage to build production-grade software.
Core foundations for building autonomous logic, backend algorithms, and user interfaces.
APIs and tooling for architecting intelligent systems, LLM agents, and prompt workflows.
Modern frameworks and tools for shipping fast, responsive, and resilient web platforms.
Data persistence layers, schema modeling, containerization, and deployment platforms.
Bilingual web application helping Pakistani students find the best university based on budget, location, and degree requirements.
Automated Open and Distance Learning application processing system for HEC with RAG, LangChain, and multi AI failover architecture.
Personal digital broadsheet newspaper featuring an autonomous four-agent editorial workflow, adaptive learning from reader habits, audio narration, and Gemini-powered Q&A.
Have an idea, project, or role you'd like to discuss? Reach out directly or fill in the form.