Amir Exir Portfolio

Professional Engineer (P.E.) and NERC Certified Reliability Coordinator with 6+ years in power systems across operation, planning, and modeling at ERCOT and LCRA. Experienced in PSS/E, GE EMS SCADA/TSM/DTS, ABB MMS and Python. Currently pursuing an M.S. in AI at UT Austin. AWS/IBM Certified.

⚡ Amir Exir’s Interactive AI Projects

Tools built to automate, simplify, and accelerate engineering workflows using AI, NLP, and power system knowledge.

ERCOT All-in-One AI Assistant

An intelligent assistant trained on ERCOT’s Nodal Protocols, Planning Guides, Interconnection Handbook, and Working Group manuals (DWG & SSWG). Designed to help engineers, planners, and developers quickly navigate ERCOT’s technical standards—covering topics like PMCR/DCP submissions, Section 6.9 generator modeling, dynamic validation, and market rule compliance.

Ask the ERCOT AI Assistant

Amir Exir's ERCOT AI DWG & SSWG Assistant

Built to answer questions from ERCOT’s Dynamic and Steady-State Working Group manuals. Supports queries about flat start case development, model validation, and NERC compliance.

Ask the ERCOT DWG & SSWG Bot

Amir Exir's ERCOT AI Planning Guides Assistant

Interact with the official ERCOT Planning Guide — get clear answers about planning procedures, updates, and compliance expectations.

Ask the ERCOT Planning Bot

Amir Exir's ERCOT AI Resource Integration Assistant

Designed to support engineers and developers with ERCOT’s interconnection process. Ask questions about data submittals, timelines, and requirements.!

Ask the ERCOT Resource Integration Bot

Amir Exir's ERCOT AI Nodal Protocals Assistant

Helps you quickly search and understand ERCOT Protocols — from market rules to operating procedures.

Ask the ERCOT Nodal Protocals Bot

Amir Exir's PSS/E API Assistant

An AI-powered Copilot-style tool that helps engineers automate common PSS®E workflows using Python. It leverages semantic search over curated technical materials and examples to generate scripts for contingency analysis, dynamic simulation, and model tuning.

Ask the PSS/E Automation Bot

Amir Exir's PSS/E Multi-Agent Automation Bot

An advanced AI-driven multi-agent assistant for automating complex PSS®E tasks using Python. It combines task planning, retreival, execution & intelligent code generation Agents in an Agent loop, and retry mechanisms to produce validated scripts for workflows like contingency analysis, load flow, and GUI-based tools. Powered by semantic search over curated PSS/E API and real examples.

Try the PSS/E Multi Agent Automation Bot

⚡ Power Fault Classifier

An interactive AI tool that predicts power system fault types using current and voltage measurements (Ia, Ib, Ic, Va, Vb, Vc). Built using Scikit-learn, Streamlit, and real system data — ideal for field diagnostics and system protection training.

Launch the Fault Classifier App

📥 Download Sample Test CSV

📎 Test dataset adapted from Electrical Fault Detection and Classification by eSathyaPrakash on Kaggle.

🔮 Hourly Load Forecast App (AEP / PJM)

📊 Dataset source: Kaggle - PJM Hourly Energy Consumption

⚠️Disclaimer⚠️:
The AI assistants provided on this site are built using public technical documentation and semantic search to help users navigate ERCOT planning guides, interconnection procedures, and scripting tasks. However, all responses are generated by AI and may contain inaccuracies or omissions.

Users are fully responsible for validating the information before applying it in operational, regulatory, or professional contexts. Amir Exir does not assume any responsibility for decisions, compliance outcomes, or damages resulting from use of these assistants.

TeslaFSDKiller: Deep Learning Project

I trained a neural network to drive autonomously in SuperTuxKart using CNNs, Transformers, and MLPs. This project, inspired by my Deep Learning course at UT Austin, showcases waypoint prediction and dynamic driving through an open-source simulator.

⚙️ PSS/E Killer GUI – N-1 Contingency Analysis

This demo showcases a Python-based GUI that performs N-1 contingency analysis on custom-built power system models. Built using Tkinter, it allows adding buses, branches, and simulating outages with real-time results.

🧠 PSS/E GUI for N-1 Contingency Analysis

I developed this Python-based GUI using the PSS/E API to automate N-1 contingency analysis. The tool visualizes buses, branches, and post-contingency violations, streamlining reliability assessments.