Two versions, one career.

2025 — Present

Vancouver, BC

ML Developer → AI Lead

CoGo · Rebranding to Lumicent — industrial Asset Risk Management (ARM) platform

• Architected an agentic copilot that acts on live platform data — assets, sensors, anomalies, alarms, maintenance and risk context — through structured tools rather than free-floating chat. The tool-grounding against the real product is the entire differentiator over a bolt-on chatbot.
• Built the production orchestration underneath it on Azure OpenAI: stateful sessions, chained responses, bounded tool loops, context compaction, and audit/persistence logging — the plumbing that keeps an agent from hallucinating its way into a stuck loop.
• Designed and shipped an automated reporting layer that turns raw vibration, temperature, anomaly and alarm telemetry into a plain-language daily briefing for each asset — what changed, what is at risk, and what needs attention — in place of a dashboard someone has to interpret.
• Extended it into multi-pass reasoning: vector retrieval over prior assessments surfaces relevant precedent, which a second model pass folds into the current analysis, giving a continuity that a single-shot response cannot produce on its own.
• Engineered for Azure OpenAI's real failure modes — rate limiting and HTTP 429s — with retries, circuit breakers and concurrency gating, and defaulted to identity-based access via Entra ID and Managed Identity over embedded API keys wherever the platform allowed it.
• Own AI planning end to end: decomposing initiatives into Azure DevOps work items, running sprint planning, and acting as the primary reviewer for the team's AI-related code and architecture.
• Present AI capabilities directly to Sales and Marketing, translating agent architecture into customer-facing value — the same translation instinct that shows up across every role on this page.

Jan 2024 — Sep 2025

Burnaby, BC

Technical Lead — Product, Cloud & Device Systems

Cannabix Technologies Inc · Publicly traded — breath-based detection devices

• Led cross-functional design changes across firmware, mechanical and electronics teams, aligning feature development with compliance requirements, customer feedback and market readiness, under direct supervision of the VP.
• Owned the architecture of a real-time AWS infrastructure — IoT Core, EventBridge and DynamoDB — powering 24/7 alcohol monitoring for customers globally, with automated test scheduling, device registration and compliance reporting.
• Built and enforced a version control and update delivery system for globally deployed devices, using S3 and cloud-hosted release tracking.
• Delivered a last-minute compliance dashboard for distributors under an aggressive deadline, shipping a working MVP in under two weeks.
• Anticipated recalibration needs and built proactive SNS/email notifications, cutting missed recalibrations by 60%.
• Took ownership of regulatory risk: designed non-editable audit trails for legal test data and engaged directly with distributor legal teams to get sign-off.
• Mentored junior developers and led the GUI codebase transition through hands-on code review, paired debugging and structured knowledge transfer.
• Partnered with leadership on product roadmap prioritization, weighing technical feasibility against resource constraints.

Aug 2021 — Jan 2024

Burnaby, BC

Senior Machine Learning Engineer

Cannabix Technologies Inc · Contract, full-time — breath-based detection devices

• Developed ML pipelines and cloud-connected software for alcohol, THC and metabolic breathalyzers — data cleaning, feature engineering, classification and the decision logic on top.
• Trained classification and regression models on sensor data and improved accuracy by mitigating environmental interference: humidity, pressure and cross-contamination.
• Built end-to-end breathalyzer GUIs in PyQt, balancing a clean field-facing UX against full sensor diagnostics underneath.
• Designed the AWS-based serverless backend — Cognito, Lambda, S3, RDS, SNS and SES — to manage devices, users and real-time test scheduling.
• Prototyped and deployed a web portal supporting distributors across North America, Australia and South Africa.
• Introduced automated recalibration workflows using dry-gas reference tests and ML-based self-correction, extending device lifespan by roughly 50%.
• Cut customer support load by more than 40% through smart on-device diagnostics and over-the-air update support.

May 2021 — Jan 2023

Remote

Senior ML Scientist / Software Engineer

Ballard Power Systems · Held concurrently with the Cannabix role — remote

• Built software that automated data reading and optimization-based EIS signal analysis, designed to be run by maintenance and sales staff with no engineering or coding background.
• Automated data cleaning, analysis and real-time visualization in Python for Ballard's air-cooled and water-cooled fuel cell stacks, used directly by experimentalists.
• Presented Ballard Power Systems as lead scientist at the International Workshop on Impedance Spectroscopy (IWIS), Chemnitz, Germany, September 2022.

Mar 2020 — Apr 2021

Burnaby, BC

Research Internship

Ballard Power Systems

• Created a numerical model in C++ and Python to analyze the effects of air stoichiometry on hydrogen emissions, and developed an ANN-based classifier/regressor on top of it.
• Used autoencoders, one-class SVM and K-means clustering in Python to detect hydrogen starvation in Ballard PEMFC stacks in real time.

Sep 2016 — Apr 2021

Greater Vancouver

Research Associate

Simon Fraser University · School of Mechatronic Systems Engineering

• Developed a virtual hydrogen sensor for fuel cell starvation detection, applying machine learning and deep learning to large datasets in Python, MATLAB and Microsoft Visual C++.
• Built a computationally efficient pseudo-2D solid oxide fuel cell model in Visual C++ and MATLAB, predicting hydrogen concentration, current density and cell temperature up to 10^5 times faster than existing methods.
• Published an in-depth review of fuel cell CFD and FEM modelling, with applications in ANSYS and COMSOL.
• Improved PEMFC maximum power density by 14% through a two-phase model coupled with a genetic algorithm.
• Ran CFD and phase-change-material optimization in MATLAB and ANSYS Fluent for a 72 kW solar-assisted absorption chiller, improving system efficiency.
• Modelled Fe2O3 magnetic nanoparticle flow in COMSOL Multiphysics and ANSYS Fluent UDFs, pioneering analysis of magnetic field effects on fluid streamlines.
• Designed net-zero energy building analysis achieving a 63% energy efficiency increase across residential and commercial cases.
• Estimated multi-physiochemical properties of SOFC electrode materials (yttria-stabilized zirconia, lanthanum strontium manganite) through pattern-search inverse optimization.
• Improved condensation heat transfer by 109% through experimental work on R600a/POE/CuO nano-refrigerant in optimized tube geometry.

Sep 2012 — Jul 2015

Tehran, Iran

Graduate Research Assistant

University of Tehran

• Worked with the Shahid Rajaei 1000 MW thermal power plant to design and optimize a hybrid dry cooling tower and solar chimney concept in ANSYS Fluent and MATLAB, raising total plant thermal efficiency by 0.5%.
• Led a team of 15 designing a residential net-zero energy building in Tehran, applying thermal mass, a Trombe wall, roller shading, flat-plate and evacuated solar collectors, fiber optics, photovoltaics and solar absorption heat pumps.
• Ran a feasibility analysis on using a Gurney flap on supercritical NASA airfoils; turbulent simulation in MATLAB and ANSYS Fluent showed up to a 50% increase in lift-to-drag ratio.
• Derived a 3D analytical model for the temperature distribution of a vertical drill embedded in soil under variable heat flux.
• Analyzed transient surface temperature behaviour for chip-and-substrate cooling in large-scale microelectronics, using ANSYS Fluent, MATLAB and Visual C++.
• Co-translated 'An Introduction to Heat Transfer' (Bergman, Lavine, Incropera & DeWitt) into Persian, with Prof. H. Shokouhmand and M.A. Bijarchi.

Ph.D. — Applied Sciences — Mechatronic Systems Engineering

Simon Fraser University · Vancouver, Canada · GPA 4.33 / 4.33

Thesis · Modelling and Diagnosis of Solid Oxide Fuel Cells (SOFC) (Supervisor: Prof. Krishna Vijayaraghavan)

M.Sc. — Mechanical Engineering — Thermal Sciences & Energy Conversion

University of Tehran · Tehran, Iran · GPA 4.00 / 4.00

B.Sc. — Mechanical Engineering

University of Tehran · Tehran, Iran · GPA 4.00 / 4.00