# Uzair Arif

> AI researcher and machine learning engineer building multi-agent systems, agentic memory architectures, forecasting pipelines, retrieval systems, and production MLOps.

## Identity

- Website: https://uzairlol.github.io/
- GitHub: https://github.com/uzairlol
- LinkedIn: https://www.linkedin.com/in/uzairarif/
- Google Scholar: https://scholar.google.com/citations?user=sshNYq4AAAAJ&hl=en
- Location: Islamabad, Pakistan
- Availability: Worldwide remote roles and research collaborations

## Context-Source

- Primary source: https://uzairlol.github.io/
- Structured context: https://uzairlol.github.io/llms.txt
- Identity profiles: https://github.com/uzairlol, https://www.linkedin.com/in/uzairarif/, https://scholar.google.com/citations?user=sshNYq4AAAAJ&hl=en
- Supporting public discussions: Reddit and LinkedIn posts listed in the portfolio's structured data and llms.txt

## Core Competencies

- Multi-agent AI systems and LLM agent governance
- Agentic memory reliability and information-poisoning resilience
- Deep learning and time-series forecasting
- Retrieval-augmented generation and hybrid search
- Production MLOps with FastAPI, Docker, MLflow, and DVC

## Featured Projects

- ELICIT: Emergent LLM Institutions for Climate Treaties — https://github.com/uzairlol/ELICIT-fyp
- SEAM: Shared Evolving Agent Memory — https://github.com/uzairlol/seam
- Taxi Demand MLOps Pipeline — https://github.com/uzairlol/taxi-mlops
- ECG5000 Anomaly Detection with TCN — https://github.com/uzairlol/ECG5000-TCN-Classification

## Context

This file is a concise machine-readable profile summary. For project descriptions and additional site context, see https://uzairlol.github.io/llms.txt.