According to the U.S. Energy Department, oil production in America increased by less than 1% during the first half of 2019—that represents a decrease of 7% YoY over the same period in 2018. Given that the gains from onshore oil drilling, and in yield and efficiency are beginning to flatten, the oil and gas industry needs to dive into another deep engineering phase to once again increase efficiency.
Houston-based oil and gas startup Quantico Energy Solutions thinks they’ve found one solution to this problem: artificial intelligence (AI).
Founded in 2012, Quantico focuses on applying AI to the key subsurface oil and gas challenges. Their core competency is using AI to increase resolution and lower cycle times for reservoir characterization. They do this by constraining the AI with physics to achieve accurate results despite sparse subsurface data and are well positioned to use AI driven subsurface prediction to lower the cost of energy exploration, development and production.
Quantico’s AI has already been applied on hundreds of wells on U.S. land and in deepwater, and according to Nathan Chang, Quantico’s director of operations, the company is now positioning itself for growth—“growth of our services into SaaS, growth in the market place of the energy value chain and growth in value for our shareholders.” Their next major advancement will be the launch of QEarth, the industry’s first real-time, high-resolution earth model.
We recently spoke to Nathan Chang, who oversees recruiting, business development, operational production, product management, and marketing for Quantico, about his experiences at the company.
What’s one unique thing that most people don’t know about what your company does?
Of any pure play AI company, Quantico works for more major oil companies around the world than any other company. Our customers include Shell, Equinor, Exxon, Conoco Phillips, and Nabors Industries.
How do you differ from your competitors?
Unlike other startups in the space, Quantico strictly focuses on the subsurface, or engineering challenges where understanding the heterogeneity of geology is most important.
How are you looking to deliver exceptional experiences for your customers? How has AWS helped you achieve that?
In multiple instances we are looking to extend the reach of Quantico’s deep learning neural networks (NN) into the client environments with seamless integration. AWS allows us to extend our models as microservices with the ability to bring the full breadth of our data science workflows into the client desktop environment through cloud integration and deployment. An easy reference is our work with Shell.
Could you elaborate some more on how AWS has helped you achieve this? What AWS services are you using?
AWS is the clear leader in terms of both cost and performance for Quantico’s demanding workloads. Big data requires massive amounts of storage, database throughput and compute power. AWS’s reliability and scalability continues to meet the demands of Quantico’s proprietary machine-learning algorithms. Among others, we are using Amazon S3, Amazon API Gateway, Amazon SageMaker, Amazon Cognito, Amazon DynamoDB, Amazon Aurora, and AWS Lambda.
What’s on your roadmap for the rest of this year and the next few years? What is the most critical initiative you’re working on now?
Specifically, we’re looking to (1) build direct plugins into client desktop software environments that leverage cloud native NN microservices (2) integrate multiple service offerings into our QEarth platform and (3) automate all of our service offerings and ensure decoupled earth modeling services are available into any end point.
from AWS Startups Blog