Causality at scale

Use your data to go beyond predictions and explanations.

About Us

Rocket Vector is an AI/ML research and deployment company. Our mission is to enable businesses and organizations to solve problems using modern causal paradigms at scale. We aim to go beyond "curve fitting" approaches to uncover profound understanding of your problem domain.

Causation over Prediction

Actionable insight is not about having predictions on outcomes but rather acting on meaningful interventions uncovered by stable causal relationships to change the outcomes.

Causal modeling does not have to be hard. With our framework, simply upload your data and our algorithms will discover robust causal relationships. With our end-to-end capabilities, we will enable you to start causal analyses in no time.

  • Discover the structural relationships.
  • Find answers to counterfactuals.
  • Identify the most efficient way to execute prevention and intervention measures to change outcomes.

With our one-of-a-kind cloud-based causal learning platform, solve your most pressing business problems at scale in minutes, not days and months.


The proof is in the pudding. Take a look at some of our demonstrations.

Our Services

We are here to help and serve.

Model Deployment

Upload your data and REST-ful endpoints will be available to start making probabilistic and causal queries.

Model Applications

If you need frontend applications to wrap around your models, we have you covered. Our world-class engineers will help you deploy customer facing applications.


We provide end-to-end consulting to solve your problems using our causal learning and inference framework.

API Preview

Too good to be true; it's easy to get started. Here's a sample of what's possible. First, upload your file and deployment specification.

              curl -X 'POST' \
                '' \
                -H "api-key: [YOUR_API_KEY]" \
                -H 'accept: application/json' \
                -H 'Content-Type: multipart/form-data' \
                -F 'param_file=@deployment-specs.json;type=application/json' \
                -F 'data_file=@data.csv;type=text/csv'

After your model is deployed, start issuing queries.

              curl -X 'POST' \
                '' \
                -H "api-key: [YOUR_API_KEY]" \
                -H 'accept: application/json' \
                -H 'Content-Type: application/json' \
                -d '{"deployment_id": "[YOUR_DEPLOYMENT_ID]", "evidence": {"sneezing": "true", "coughing": "false"}}'

All models are Bayesian Belief Networks (BBNs). We currently support learning and inference for discrete and Gaussian BBNs.

If you want to go under the hood, please refer to our API and SDK documentations below.

  • API Documentation: Use our API to learn causal models in the cloud
  • SDK Documentation : Use our SDK to learn causal models on-prem
  • py-bbn Documentation : Use py-bbn to derive actionable insight through probabilistic, interventional and counterfactual reasoning (for categorical data)
  • py-scm Documentation : Use py-scm to derive actionable insight through probabilistic, interventional and counterfactual reasoning (for Gaussian distributed data)


Email us at to start an initial engagement. We are currently in private preview mode. We welcome and look forward to your support and help.