Explore our Use Cases:
SUSTAINABILITY
Using Reinforcement Learning for Water Treatment
Challenge
Traditional water treatment facilities rely on fixed processes that waste chemicals, incur high operating costs, and negatively impact the environment.
Outcome
RL Core Technologies leveraged Reinforcement Learning to develop an automated control system – with human oversight – that learns, adapts, and optimizes water treatment in real time.
Impact
RL Core's technology has boosted efficiency by 10–20% annually, saving thousands in resources and maintenance, and is now active in three water and wastewater plants across Alberta and British Columbia.
+20%
Efficiency Boosted
RL Core: An Amii Success Story
RL Core Technologies is a startup founded by two current Amii fellows.
Martha White
Martha White is also co-founder of RL Core, as well as an Amii Fellow, a Canada Research Chair in Reinforcement Learning and a member of The Royal Society of Canada's College of New Scholars, Artists and Scientists. Her work focuses on developing adaptive autonomous agents that can work in real-world industrial settings
Adam White
Adam White is the co-founder of RL Core Technologies. He is also an Amii Fellow, a Canada CIFAR AI Chair, and a Professor of Computing Science at the University of Alberta. His work and research focuses on how to leverage artificial intelligence to replicate human-level intelligence in physical and simulated agents.
More Reinforcement Learning Success Stories
HEALTH CARE
Transforming limbs and lives with RL
CHALLENGE
Traditional prosthetic limbs lack the intuitive control needed to feel like a natural part of the body. They rely on "pre-programmed" movements that require the user to perform awkward, unnatural gestures to trigger a specific grip or step. Because these limbs cannot adapt to the user, they feel like a tool rather than a part of the body.
OUTCOME
The BLINC Lab uses Reinforcement Learning to transform prosthetics from rigid tools into intuitive body extensions that learn and adapt in real-time. The limb uses RL to constantly listen to the user’s muscle signals and track its own position.
IMPACT
By learning and adapting to the user, the limb stops being a "device" that a user has to operate and starts being a part of who you are. The real impact is the freedom to simply live life without having to think about the technology making it happen.
INDUSTRIAL NOISE POLLUTION
Tackling noise for safer working conditions
CHALLENGE
Extreme workplace noise causes major health issues. Prolonged exposure to high-decibel environments leads to irreversible hearing loss, chronic worker stress, and dangerous levels of fatigue. This also leads to higher healthcare costs, lower productivity, and strict regulatory penalties.
OUTCOME
Using Reinforcement Learning, we achieved noise reduction of 12–15 dB at target frequencies (≈94–97% less sound energy). The system automatically adjusts in real time as conditions change (temperature, humidity, panel position), is effective for single and dual-frequency noise, and scalable to more frequencies.
IMPACT
By eliminating nearly 97% of harmful sound energy, we protect employees from life-altering conditions like hearing impairment and hypertension, ensuring a healthier, more focused workforce.
ENERGY
Optimizing gas wells to reduce emissions
CHALLENGE
In many upstream oil & gas operations, gas wells naturally struggle with "liquid loading"—essentially, water pools at the bottom and chokes off production. While Plunger Lifts are the standard tool to clear these fluids, they are often a victim of "set it and forget it" management. Because field technicians are stretched thin, they often rely on static timers or rigid rules that don't adapt to a well's changing needs. This leads to a costly ripple effect: lost production, increased methane emissions from manual venting, and premature equipment failure. increase, and operating costs are unnecessarily high.
OUTCOME
We have developed a specialized model that manages and adjusts hundreds of wells simultaneously. Our system automatically adjusts the two most critical "trigger" points for each plunger lift every few days—a level of precision that is impossible for a human team to achieve across a large fleet.
IMPACT
We maximize the well’s natural pressure to lift fluids, significantly reducing the need for costly external power and manual site visits. Automated, precision-timed cycles also reduce mechanical stress on equipment and extend the profitable life of your existing wells. By maximizing the life of these assets, we reduce the environmental impact of new drilling and ensure a more responsible, low-emission energy future.
SEMICONDUCTOR
Automating complex circuit design
CHALLENGE
The world is running on more chips than ever. Today’s electronics are so complex that even a simple circuit can take an expert engineer weeks of manual "trial and error" to perfect. With costs skyrocketing and a global shortage of skilled designers, the traditional way of building hardware is hitting a wall. We simply cannot keep up with the demand for new technology using old, manual methods.
OUTCOME
Our new AI platform automates complex design, taking a basic blueprint and automatically finds the perfect size and settings for every component. It slashes workloads from weeks down to hours, letting companies innovate at the speed of software.
IMPACT
Currently, high-end chip design is so expensive that only a few giant corporations can afford it. By automating the most complex parts of the process, we are empowering startups and local innovators to create specialized technology that makes essential devices like smartphones and life-saving medical devices more affordable for everyone.
Ready to explore what Reinforcement Learning could do for your business?
Book a meeting with Amii
At Amii, we've built deep expertise in identifying ways to address complex problems using AI. Now, we’re ready to bring that experience to your business.
Fill out the form below, and a member of our team will get in touch.
