🎙️ Revolutionizing Gas Leak Detection with Visual AI & Machine Learning

In this episode of The Empirical Energy Podcast, host Mark Smith sits down with Jae Yoon Chung, Machine Learning Engineer at Clean Connect, to explore how visual AI is transforming gas leak detection in the energy industry.

Jae breaks down the real-world challenges of detecting methane and gas leaks using vision-based models — especially in harsh outdoor environments with wind, rain, snow, and limited edge-device compute. He introduces a breakthrough approach called channel stacking, a method that captures gas movement using just three consecutive frames to dramatically improve detection accuracy while reducing computational load and false alarms.

The conversation goes beyond theory, offering a behind-the-scenes look at how AI models are trained, optimized, and deployed at the edge — and where the technology is headed next. From edge computing to large language models (LLMs) and object-level incident classification, this episode highlights how AI, blockchain, and verification are reshaping the future of global energy markets.

⚡ If you work in energy, AI, emissions monitoring, or industrial technology, this episode is a must-listen.

⏱️ Episode Chapters

00:00 – Introduction to The Empirical Energy Podcast

00:57 – Meet the Guest: Jae Yoon Chung from Clean Connect

01:39 – Machine Learning Challenges in Energy Environments

03:44 – Innovations in Visual Gas Leak Detection

07:29 – Technical Deep Dive: Channel Stacking Explained

14:23 – The Future of Visual AI & LLM Integration

18:53 – Final Thoughts & Call to Action

🎧 Listen & Watch

▶️ YouTube: https://youtu.be/WOejlQSdr_g

🎙 Apple Podcasts: https://podcasts.apple.com/us/podcast/the-empirical-energy-podcast/id1822839881

💬 If this episode sparked new ideas or questions, join the conversation.

👉 Subscribe, rate the show, and share this episode with someone working in energy, AI, or climate tech.

👉 Drop a comment and tell us: Where do you see the biggest opportunity for AI in energy today?

#EmpiricalEnergyPodcast #VisualAI #MethaneDetection

#MachineLearning #EdgeAI #EnergyTech

#ClimateTech #IndustrialAI #ComputerVision

Uncovering the Future of Methane Emissions Management with Thomas Fox

In this episode of The Empirical Energy Podcast, host Mark Smith welcomes Thomas Fox, President of Highwood Emissions, to unpack the mechanisms reshaping global energy markets.

The conversation explores the industry’s shift from “alternative energy” toward measured, verified energy as the new gold standard—powered by empirical data, evolving regulations, and emerging verification frameworks. Thomas shares his journey from PhD research in methane emissions to building Highwood Emissions and launching the Emissions Intelligence Platform (EIP), a software solution designed to help operators create measurement-informed inventories at scale.

Together, they dive into the realities of methane reporting: reconciliation challenges, OGMP Level 5 requirements, EPA and Colorado regulatory approvals, and why harmonization across global reporting frameworks remains one of the industry’s biggest hurdles. The episode closes with a forward-looking discussion on how collaboration, technology, and credible data may finally unlock real market incentives for low-methane energy.

🎧 If you work in energy, ESG, emissions reporting, or climate compliance, this episode offers a clear look at where the industry is heading—and what it will take to get there.

⏱️ Episode Chapters

00:00 Introduction to the Empirical Energy Podcast

00:55 Meet Thomas Fox: Highwood Emissions

02:51 The Journey to Methane Mitigation

04:54 Building Measurement-Informed Inventories

08:27 Challenges and Innovations in Methane Reporting

18:34 The Future of Emissions Reporting

25:02 Closing Thoughts and Call to Action

✔️ Listen to the full episode to understand how methane measurement is evolving

✔️ Subscribe to the Empirical Energy Podcast for more conversations at the intersection of energy, AI, and climate policy

✔️ Share this episode with colleagues navigating methane reporting, OGMP, or ESG compliance

✔️ Join the discussion—what do you think will finally drive harmonized emissions reporting?

#EmpiricalEnergy #MethaneEmissions #EnergyTransition #OGMP #ClimateReporting #EnergyMarkets #OilAndGas #ESG #CarbonAccounting #Decarbonization #ClimateTech #EnergyPolicy #LowMethaneGas #Sustainability

In this episode of the Empirical Energy Podcast, host Mark Smith sits down with Zach Orlin of Clean Connect.ai to explore how digital twins, AI, blockchain, and verification technologies are redefining how energy is measured, monitored, and managed.

The conversation dives deep into Luminary, a digital twin platform that enables teams to virtually map energy sites, plan monitoring equipment placement, and integrate real-time data into operational models before anything is deployed in the field. Zach explains how Luminary connects with Promax process simulation models, remote monitoring systems, and verification frameworks to reduce costs, improve efficiency, and eliminate blind spots in energy operations.

This episode also explores what’s next—from augmented reality overlays to smarter, more empirical energy markets where measurement, transparency, and real-time intelligence become the new standard.

If you’re working in energy, infrastructure, emissions monitoring, or digital transformation, this episode offers a clear look at how legacy systems are being replaced by data-driven, verifiable solutions.

⏱ Episode Chapters

00:00 Introduction to Empirical Energy

00:44 Meet Zach Orlin (Clean Connect.ai)

02:23 Digital Twins & the Luminary Platform

05:05 Virtual Site Mapping & Monitoring Design

11:11 Advanced Capabilities & Future Applications

18:13 Empirical Energy’s Broader Impact & Wrap-Up

🎧 Subscribe for conversations on energy verification, AI, blockchain, and the future of global energy markets.

#EmpiricalEnergy #DigitalTwins #EnergyTechnology #EnergyInnovation #AIinEnergy #BlockchainEnergy #EnergyMarkets #RealTimeMonitoring #EnergyManagement #CleanTech #IndustrialAI #EnergyTransition #OperationalIntelligence