Hi, my name is

Mohammad Javad Raei

Scientific Software Developer

I build agentic AI systems that reason, retrieve, and act reliably enough to trust in production.

About Me

I'm an AI engineer focused on agent design: how tools get exposed to a model, how state persists across a run, and how you know whether a system actually worked. I build multi-agent systems, retrieval-grounded architectures, and the eval harnesses that measure reliability rather than just demo well.

Professionally, I develop engineering software in the oil and gas industry. Outside of work, I build agentic AI tools for scientific research, work that has earned me four prizes across two hackathons.

When I'm not coding or experimenting, you'll find me behind the camera. I'm passionate about photography and video editing.

Mohammad Javad Raei

Experience

Software Developer - Full-Stack

Opla Energy

Dec 2024Present · Calgary, Alberta, Canada

  • Design and maintain REST APIs and cloud services on AWS (EC2, S3, Lambda) for telemetry ingestion, software distribution, and version management, giving direct visibility into how shipped software behaves in the field.
  • Build and maintain CI/CD pipelines for automated testing, build, and release.
  • Maintain end-to-end test coverage across numerical, integration, and regression paths (pytest), so refactors stay safe to make quickly and behaviour remains verifiable as the system changes.
  • Build real-time data processing and interactive analytics interfaces for scientific calculations, used daily by non-technical domain users (Python, QML, Plotly).
  • Develop automated reporting modules that generate dynamic outputs from live data, removing manual preparation steps and enforcing consistency across reports.
  • Refactor legacy codebases and profile computational bottlenecks to improve application speed and scalability, holding a modular, testable, cross-platform architecture.
  • Own features end to end across design, backend, interface, release, and monitoring, with no handoff at the prototype boundary.
  • Work directly with domain experts to convert underspecified requests into scoped technical designs, making pragmatic calls about what to cut to ship.
PythonQMLPlotlyREST APIsAWS (EC2, S3, Lambda)CI/CDpytestGitLinux

Air Emissions Advisor - ML Engineer

Process Ecology

Sep 2023Dec 2024 · Calgary, Alberta, Canada

  • Built and validated machine learning models for production forecasting and carbon credit estimation, including XGBoost regressors and classifiers, SARIMA for expected-behaviour baselines, K-Means for behavioural segmentation, and PCA for dimensionality reduction on wide feature sets (Python, Scikit-learn).
  • Developed unsupervised anomaly detection over high-volume operational time-series using autoencoders, surfacing abnormal behaviour without labelled training data and separating genuine signal from routine variation, then tracking detection performance over time.
  • Ran the full ML lifecycle end to end: feature engineering, training, cross-validation, hyperparameter tuning, structured error analysis, benchmarking against established baselines, and quantifying uncertainty to define when outputs could be acted on.
  • Led back-end development of an emission quantification platform in C# using object-oriented design, with calculations built to be reproducible and traceable to the evidence behind them under external audit.
  • Engineered ETL pipelines in Python and SQL across disconnected client systems: discovering and accessing new data sources, then ingesting, cleaning, verifying integrity, and making data model-ready (Pandas, NumPy).
  • Wrote complex multi-source SQL for analysis, and automated repetitive data workflows with Selenium to improve operational accuracy.
  • Mentored a small engineering team: running standups, scoping and assigning work, reviewing code, and holding delivery timelines against client commitments.
  • Prepared model documentation, regulatory reports, and presentations for external review, and presented findings to both technical and executive stakeholders.
PythonScikit-learnXGBoostSARIMAAutoencodersC#SQLPandasNumPySeleniumGit

Education

Master of Science in Chemical Engineering

University of Calgary

Sep 2021Sep 2023 · Calgary, Alberta

  • Applied design of experiments and response surface methods to optimize experimental conditions, with data pipelines for processing, regression modelling, and analysis.

B.S. Chemical Engineering, Minor in Mechanical Engineering

Sharif University of Technology

Sep 2016Jun 2021 · Tehran, Iran

Skills

Agent Engineering

LLMsVLMsClaudeOpenAIGeminiHugging FaceOllamaMulti-Agent OrchestrationTool/Function CallingRAGFAISSVector SearchEmbeddingsEval HarnessesLLM-as-JudgeHuman-in-the-Loop GuardrailsProgressive DisclosurePrompt DesignPrompt CachingContext-Window ManagementCost & Latency OptimizationStreaming UXNLP

ML & Data

XGBoostscikit-learnSARIMATime-Series ForecastingK-MeansPCAAutoencodersAnomaly DetectionUncertainty QuantificationBenchmarking & Error AnalysisNumPyPandasSciPyOpenCVPlotlyETL Pipelines

Languages

PythonTypeScriptJavaScriptC#SQLBashHTMLCSS

Frameworks & Libraries

Google ADKLangGraphLangChainPydanticChainlitFastAPIReactNext.jsNode.jsExpressREST APIsTailwind CSS

Databases

PostgreSQLMongoDBSQLiteRedis

Tools & Platforms

GitDockerAWSCI/CDpytestVercelLinuxSelenium

Awards & Recognition

4th Place: Prize by Lila Sciences

2025 Hackathon on LLMs for Materials Science & Chemistry

Sep 2025

CrystaLenz ranked 4th among top-scoring projects in a global competition with over 1,000 participants across 100+ teams globally.

Theia Scientific Award

2025 Microscopy Hackathon

Nov 2025

Awarded for NanoRange, an AI-powered nanoparticle analysis platform using multiple specialized agents to detect, measure, and visualize nanoparticles from microscopy images.

DENS Mystery Award

2025 Microscopy Hackathon

Nov 2025

Second award for NanoRange at the same hackathon, recognizing the platform's innovative approach to automated microscopy analysis.

1st Place: SPE Datathon 2024

Society of Petroleum Engineers (SPE)

Oct 2024

Designed and implemented ML models (Random Forest, SARIMA) to forecast carbon emissions and support CCUS decision-making from large-scale climate data.

Alberta Innovates

University of Calgary

Sep 2021

This funding was provided for my master's studies to design novel materials for carbon capture in oil and gas industries.

Publications & Patents

NanoRange: A Modular Framework for Automated Cryo-EM Particle Detection and Morphometric Analysis

S. A. Golsorkhi, M. J. Raei, F. J. Alvarez, M. Mamak

Microscopy and Microanalysis

Jul 2026

From Knowledge to Action: Outcomes of the 2025 Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry

A. Roy, K. Shen, A. MacBride, A. Oladipupo, M. Taskeen, W. Treyde, ...

arXiv e-prints

May 2026

One-Step Synthesis of Nitrogen/Sulfur Codoped Graphene/MnO2 Film as a High-Performance Supercapacitor

M. J. Raei, M. Trifkovic, E. P. L. Roberts, G. Natale

ACS Applied Energy Materials

Jul 2025

(Pending) Methods of Forming Exfoliated Graphene-Based Materials, Exfoliated Graphene-Based Materials, and Uses Thereof

S. Pal, M. Trifkovic, M. J. Raei, E. P. L. Roberts, G. Natale

US Patent

Oct 2023

Get In Touch

I'm always open to new opportunities and interesting projects. Feel free to reach out if you'd like to connect.