Decision intelligence in hours, not months

Setting up effective, reliable systems for sequential decision making is a time-consuming, error-prone task. Especially when you’re trying to use cutting-edge methodology. Prophet offers you a proven, research-based framework for model predictive control. Simplify the process of setting up, solving, and monitoring MCP problems. Impact your system’s performance and effectiveness fast & reliable. Make better decisions faster.

How it works

1. Upload Dataset: Begin by uploading your dataset to the prophet API, providing your historical dataset.
2. Create Agent: Following the data upload, create an 'Agent', by defining the optimization goal and constraints.
3. Train Agent: The next step involves training your Agent on the uploaded dataset, allowing it to understand patterns, learn from them and make accurate future predictions.
4. Get optimized decisions: Once trained, the Agent is capable of generating optimized decisions, delivering high-performance solutions to your unique problems, thus enabling you to make better, faster decisions.
5. Routine Retraining: To safeguard against distribution shifts in your environment, Prophet facilitates routine retraining of your Agent. This ensures that your model stays up-to-date and maintains its accuracy and effectiveness over time.

Key Features

• Proven, Reliable Methodology: Prophet uses well-established methodologies that have been validated by a community of researchers and practitioners across a wide range of industries. These methodologies have been tested for robustness and have shown to consistently yield reliable results.
• Cutting-Edge Research: Built on the foundation of the latest research in model-based RL, Prophet continuously integrates innovative advancements in the field of model predictive control. This means that you always have access to the most advanced, reliable solutions for your MCP problems.
• User-Friendly Interface: Prophet’s intuitive, easy-to-navigate interface makes it simple for both experts and beginners to configure and manage their MCP systems effectively. This translates into less time learning the software and more time improving your decision-making processes.
• Continuous Learning and Adaptation: With Prophet, your MCP system will continue to learn and adapt over time, improving its decision-making capabilities. It uses machine learning algorithms to learn from historical data and adjust its models for future predictions.

Using Prophet, you can streamline your processes, enhance performance, and make better, faster decisions. Its feature-rich environment is designed to support both your immediate and long-term sequential decision-making needs. Start leveraging Prophet today and experience a significant transformation in your model predictive control tasks.

Getting started

We appreciate your interest in Prophet, the research-based framework for model predictive control. As we are currently in the development phase, we are keen on onboarding early adopters who are willing to test and provide invaluable feedback on our system. This will ensure that Prophet is fine-tuned and capable of meeting user needs efficiently upon full launch.

Join the Early Access Waitlist: Prophet is currently under development and we are offering a select group of users the opportunity to be a part of our early access program. Please click on the 'Join Waitlist' button to access the applicaiton form.