TeraBundle Analytics

We maintain computational artifacts of academic research, allowing rapid development of proof-of-concepts.
Let us help you practice Reproducible Research!

About Us

We are a startup dedicated to provide statistical and computational know-how to accelerate research. We work with you to publish your data in both user and machine friendly ways (FAIR) and your algorithms as runnables, via an intuitive and secure web interface that reviewers can access anonymously. We fast track the review process by providing reviewers with access to data that goes well beyond the supplementary, to highly interactive interfaces that simplify cross-validations and finally, consultancy services that improve the statistical and computational rigour of the publication. This makes your publication more reproducible - a criterion that is increasingly becoming important.

Computational approaches are now ubiquitous in science, but technical aspects of the associated tools change frequently and can be a significant bottleneck in the pace of publications. Science is now collaborating with engineering more than ever before - we let you focus on the Science. Individual researchers and labs do not always have the dedicated resources necessary to keep up with the increasing demands of reproducible research. Here are the primary issues we address:

Your data and tools, organized in your way - always ready to run!

Let us take a step back. We as researchers spend a lot of time organizing our data, trying various tools to perfect our analysis, but can capture only part of this in the published manuscripts and their supplementals. The rest is distributed as experience among the participants. Researchers often struggle to pull up the details, or re-run an analysis later as participants migrate and experience fade over time. Although the scientific acheivement is kept well recorded in publications, the operational insights do not get reused.

If those data files, and their associated programs were carefully managed by a third party service so that the analysis could be re-run later by us or shared as runnables to others, a significant roadblock in reproducibility can be removed. A lot of arguments with fancy figures would not be as effective as a simple validation that a user can do with their own data. Expecting researchers to manage this is not practical, they neither have the time nor the incentive as per the existing system. Graduate students and Postdocs can save time and focus more on their research.

Research Engineering: Reusable means Sustainable

Growing complexity of experiments dictate that researchers cannot afford to write throw-away code anymore. Published article is no longer the only formal deliverable of research. Shallow implementations that do not model the problem are neither extensible nor testable. Software engineering can help, but is challenging as the requirements are not clear upfront. Writing extensible research software that is testable is hard, extremely time consuming and requires fair amount of experience. No matter how good you become in managing this process, it does not differentiate your contribution as a researcher. We have proven experience in writing extensible research software covered by industry standard testing methodologies. Publish with us and we keep you running.

Demo in Action

How it Works

We work with you to get your data and algorithms interactive online. Your data can be in any standard or textual format. Your algorithms can be in any popular programming language. We convert them into suitable computable formats, called bundles and we host them on our servers, so that anybody can play these bundles. New data and algorithms get integrated in a seamless manner.


We work with you to host your data and algorithms in bundles, so that they can be easily verified or used by others. You can host with before or after the publication. We provide special access during the review process so that reviewers can anonymously acess the data and play the associated bundles. Reviewers get a live perspective of your work, which can expedite the review process. Post publication, you can also keep your bundle updated with continous feedback which snowballs your research impact.


You can play (run) the available bundles for running analysis online. You pay a fee for running bundles, just like the we pay for talk time on our phones. The cost of running a job is calculated based on its parameters and is shown to the user prior to running a job. The user can add funds to run jobs. A job is runnable only if the user has sufficient funds to run the job. Fund is deducted only if the job ran successfully. All new users start with a free US$50 worth of funds to explore jobs.


You earn a royalty every time your bundles are run. Only bundles under the advanced plan are eligible for royalty. The royalty details will depend on the particular bundle, its complexity and its estimated demand. At present, we work out the agreement details with the individual users.

Contact Us

Please don't hesitate to contact us if you need some help. You can also reach us at contact@terabundle.com

Please fill out all required fields