ENGINEERING TOOLKIT
AELab analysis suite
Run contingency studies, review dynamic simulation results, and prepare reports with Python tools for PSS/E and TARA PowerGEM.
Python · PSS/E · TARAExplore the toolkitAMIR EXIR, P.E. / POWER SYSTEMS & AI
Power meets
intelligence.
From the networks that power our world
to the intelligence that helps us understand them.
NuclearGasHydroWindSolarBattery storage
- Generation + storage
- Step-up
- Transmission
- Step-down substation
- Distribution
- Loads
EXPLORE THE INTERSECTION
A study in connection, energy, and flow.
AUSTIN, TEXAS ENGINEERING WITH PURPOSEENTER THE PORTFOLIO
01
Transmission planningModels, studies & interconnection
02
Grid operationsReliability & real-time engineering
03
Engineering automationPython, PSS/E & TARA PowerGEM
04
AI engineeringKnowledge systems, forecasting & ML
THE ENGINEER BEHIND THE WORK
Grounded in the field.
Driven by possibility.
I’m Amir Exir, P.E. I connect transmission planning and grid-operations experience with software engineering and graduate AI training.
PHYSICAL SYSTEMS. INTELLIGENT SOFTWARE.Two disciplines.
Two disciplines.
One perspective.
Engineering judgment shapes the question.
Software and AI expand what we can build.
P.E. Licensed Professional EngineerNERC Certified System OperatorUT Austin MSAI · Graduating
SELECTED WORK / THE BUILD LOG
Ideas, made
into instruments.
Explore engineering tools, AI knowledge agents, predictive models, and interactive dashboards—from power system studies to document research and market analytics.
Showing all selected projects.
6 projects
INTERACTIVE EXPLORATION
Grid Atlas
Locate transmission lines, substations, and generating facilities using public infrastructure data for U.S. grid regions and Canada.
Geospatial · Public grid dataExplore the map
KNOWLEDGE & RETRIEVAL
ERCOT knowledge agents
Research ERCOT planning, interconnection, and operating requirements with AI-assisted document search and source citations.
RAG · Retrieval · PythonExplore technical assistantsGRID ANALYTICS
ERCOT intelligence dashboard
Load forecasting, renewable generation, market pricing, and outage views in one interactive dashboard.
ERCOT API · Forecasting · StreamlitView the dashboardAPPLIED MACHINE LEARNING
Power grid GNN predictor
Explore graph neural networks for voltage and thermal alarm prediction in simulated grid scenarios.
PyTorch Geometric · GNN · PandapowerExplore the modelAUTOMATION CASE STUDY
Market intelligence agent
An experimental workflow connecting market data, model evaluation, Telegram delivery, and paper trading.
n8n · Machine learning · AutomationSee the workflowNo matching projects.
Try a different search or explore all selected work.
Professional Summary
An engineer’s perspective.
A builder’s mindset.
Amir Exir, P.E.Power systems + AI engineering
I am a Professional Engineer and NERC Certified System Operator with a background in electrical and computer engineering, experience at ERCOT, LCRA, and PEC, and I am completing UT Austin’s M.S. in Artificial Intelligence program. My work combines transmission planning, grid operations, and energy management system (EMS) applications with software and AI engineering.
I build Python applications, retrieval-augmented knowledge agents, and machine learning models for technical research, forecasting, classification, and automation. Power systems provide the foundation for much of this work; additional projects explore language-model fine-tuning, market analytics, and automated reporting.
- PSS/E and TARA PowerGEM workflow automation for planning studies and reporting.
- Technical document search across ERCOT protocols, guides, working group manuals, resource integration procedures, and revision requests.
- Machine learning applications for load forecasting, fault classification, power grid alarms, and market analytics.
Credentials
Engineering foundations.
Graduate depth in AI.
Professional licensure and system operator certification, alongside graduate education in electrical and computer engineering and advanced study in artificial intelligence, machine learning, and cloud computing.
Professional Engineer
Texas P.E. license with power systems and transmission planning focus.
Verify credential
NERC Certified System Operator
Reliability Coordinator certification and real-time grid operations background.
Discuss grid operations experience
UT Austin M.S. in Artificial Intelligence Degree candidate · Graduating
Graduate training in machine learning, optimization, deep learning, AI ethics, and applied intelligent systems.
View AI badge
M.Eng. Electrical & Computer Engineering
Lamar University, 2020. Master of Engineering in Electrical and Computer Engineering.
AWS Cloud Practitioner
Certification in AWS cloud fundamentals.
View credentialDownloads
Resume, PSS/E automation certificate, IBM AI certificate, machine learning certificate, and AI course badges.
Explore course badges & certificates
Experience
Transmission planning
and grid operations.
Planning, operations, advanced applications, and training experience across PEC, LCRA, and ERCOT.
Transmission Planning Engineer 4
Conduct transmission planning studies, evaluate projects, and analyze system reliability using power system models.
Transmission Planning / EMS Advanced Applications Engineer
Worked across planning studies, EMS advanced applications, GE EMS SCADA/TSM/DTS workflows, modeling support, and engineering automation.
Resource Integration / Operations Training / Real-Time Engineering
Supported ERCOT resource integration, operations training, real-time engineering, model requirements, and market/reliability procedures.
Associate / Substitute Teacher
Classroom teaching and instructional support, with an emphasis on explaining technical concepts to students with varied backgrounds.
Education
Electrical engineering.
Artificial intelligence.
A foundation in electrical and computer engineering, followed by graduate study in power systems and artificial intelligence.
M.S. in Artificial Intelligence Degree candidate · Graduating
Graduate study in deep learning, machine learning, optimization, and AI ethics, with projects in computer vision and autonomous racing.
M.Eng. in Electrical & Computer Engineering
Coursework in power system stability and control, protection, robotics, programmable logic controllers, computer networks, and cyber-physical systems.
B.S. in Electrical & Computer Engineering
Foundations in power systems, electrical machines, protection and relays, electronics, programming, and computer architecture.
Flagship Tool
AELab.
Built for the study workflow.
AELab brings PSS/E and TARA PowerGEM study tasks into a Python desktop application. It supports contingency analysis, dynamic simulation, model validation, results visualization, IDV generation, and transmission project reporting.
01 / AELAB · CONTINGENCY ANALYSIS
One circuit opens.
The whole network responds.
Trace the connection between topology, power redistribution, and engineering review.
Study capabilities
- Run batch contingency analysis and report thermal overloads, voltage violations, and voltage deviations.
- Process dynamic simulation outputs, run events, summarize channels, and generate plots.
- Compare PSS/E cases and power flow, dynamic, and contingency results to support model validation.
- Visualize TARA PowerGEM TRLIM and contingency results for transmission upgrade and hosting-capacity review.
- Generate IDV files from line impedance and rating data for model updates.
Python
PSS/E
TARA PowerGEM
GUI automation
Planning studies
Interactive Grid Map
Explore transmission
and generation.
Explore transmission lines, substations, and power plants on an interactive map of U.S. grid regions and Canada. Search for facilities, filter infrastructure by region, and review available asset and source information.
Data & AI Solutions
Technical knowledge.
Easier to navigate.
Retrieval-augmented AI assistants for technical document research. Search ERCOT procedures and PSS/E API references, retrieve relevant passages, and review answers alongside source citations.
KNOWLEDGE, WITH A SOURCEFrom information
CONCEPTUAL RETRIEVAL FIELD
From information
to understanding.
Retrieve the passage. Preserve the context.
Make the evidence visible.
ERCOT All-in-One Knowledge Agent
Research ERCOT requirements for planning studies, modeling, resource integration, and market operations. Search across Nodal Protocols, Planning and Operating Guides, Resource Integration materials, DWG/SSWG procedures, Other Binding Documents, and revision requests.
Review related procedures and rule changes with section and page citations and links to ERCOT source documents. Where the source evidence establishes document status, the assistant distinguishes effective requirements from proposed or future changes. Revision coverage includes NPRR, PGRR, NOGRR, OBDRR, RRGRR, VCMRR, COPMGRR, LPGRR, RMGRR, SMOGRR, CMGRR, and SCR.
Browse indexed documents and revision status
Open the ERCOT assistant to browse indexed documents and their source references.
ERCOT
RAG
Streamlit
Engineering QA
ERCOT DWG & SSWG Knowledge Agent
Answers questions from ERCOT Dynamic and Steady-State Working Group manuals, including flat-start case development, model validation, and compliance support.
ERCOT Planning Guide Knowledge Agent
Searches ERCOT planning procedures and requirements for faster navigation of standards, revisions, and study expectations.
ERCOT Resource Integration Knowledge Agent
Supports interconnection process questions around data submittals, timelines, responsibilities, and public documentation requirements.
ERCOT Nodal Protocols Knowledge Agent
Knowledge agent for ERCOT protocol research across market rules, operating procedures, settlement topics, and technical requirements.
PSS/E API Knowledge Agent
Find PSS/E API references and scripting examples for power flow, contingency analysis, and dynamic simulation.
PSS/E Multi-Agent Automation System
Develop PSS/E automation scripts with coordinated AI assistants for task planning, API lookup, code generation, execution, and error handling.
Professional-use note: AI-generated answers require engineering review. Verify citations, document status, and effective dates against official sources before using an answer in a study, operating procedure, or compliance decision.
Applied ML
Machine learning.
Practical applications.
Explore forecasting, classification, graph neural networks, and language-model fine-tuning. Applications include power grid analysis, energy demand forecasting, medical question answering, and interview preparation.
LEARNING FROM CONNECTIONSEvery connection
CONCEPTUAL NEURAL FIELD · NOT STUDY DATA
Every connection
holds a signal.
Local features. Network structure.
Models that learn from both.
Technical White Paper
Practical Machine Learning Methods for Power-System Operations and Planning (2025)
A technical white paper exploring practical applications of machine learning for electric power systems, with a focus on forecasting, classification, graph learning, and retrieval-based engineering assistants.
- Short-term load forecasting with XGBoost
- Fault classification using waveform features
- Graph Neural Networks for voltage and thermal alarm prediction
- Retrieval-Augmented Generation (RAG) assistants for ERCOT engineering documentation
White paper
XGBoost
GNN
RAG
ERCOT Load Forecast Dashboard
Review ERCOT load forecasts, renewable generation, real-time market prices, and resource outages in one dashboard. Follow industry news and regulatory updates, with reports available through Telegram.
ERCOT API
Load forecasting
Grid analytics
n8n
02 / LOAD FORECASTING · TEMPORAL PATTERNS
Patterns behind us.
Possibilities ahead.
Follow the shape of demand from historical patterns, through the forecast origin, into a changing horizon.
Power Grid GNN Predictor
Graph neural network system for simulated power grid scenarios that predicts voltage violations and thermal overloads across buses and transmission lines using GCN, GAT, GIN, and Graph Transformer architectures.
PyTorch Geometric
Pandapower
Grid topology
Classification
Power Fault Classifier
Explore fault type classification from current and voltage measurements in a Scikit-learn and Streamlit application for protection training and diagnostic research.
03 / POWER FAULT CLASSIFIER · SIGNAL INTELLIGENCE
Every disturbance
leaves a signature.
Six current and voltage channels. A focused signal window. Features that connect electrical behavior to classification.
Hourly Load Forecast App
Explore hourly load forecasts for AEP/PJM using historical demand data and interactive model analysis.
TinyLlama Medical Q&A Fine-Tuning
LoRA fine-tuning project using TinyLlama-1.1B-Chat and MedQuAD, with ROUGE-based evaluation and resource-aware training techniques.
AI Interview Assistant
Practice behavioral and technical interviews with an assistant that draws on résumé content and STAR examples.
Automation Case Study
Market data analysis
and paper trading.
An experimental stock and cryptocurrency application that compares forecasting models, summarizes market data, and supports Alpaca paper trading. Reports can be requested through Telegram or delivered on a schedule using n8n and calendar integrations. This demonstration does not provide financial advice.
Analysis and reporting
- Collect market data and evaluate trading workflows through Alpaca paper trading.
- Compare, validate, and rank forecasting models before publishing signals.
- Request individual analyses or schedule reports through Telegram.
- Save report versions and deliver summaries through Telegram and Google Calendar.
Personal
Beyond the desk.
Around Austin.
I enjoy biking and paddleboarding around downtown Austin, and I make time for the live music culture that makes Austin, Texas, feel like home.
Contact
Good work starts
with a conversation.
I welcome conversations about power systems, AI engineering, and software development. Contact me to discuss engineering automation, knowledge agents, machine learning, analytics, or solutions architecture for energy and other technical applications.