Tonic AI

What do you like best about Tonic.ai?

Tonic takes the guesswork out of providing data for engineers to use during debugging etc. We wrote a simple automation that runs Tonic every day, and as a result our developers have up-to-date info to work with that is largely set-and-forget.

What do you dislike about Tonic.ai?

Some of the security decisions are outdated. For example, Tonic requires that you provide an AWS access and secret key to log in to an AWS environment, which is a deprecated process that Amazon recommends against. These keys can be exfiltrated and used by attackers and require additional security controls on our end to protect against misuse.

Using a cross-account AssumeRole is what I would expect instead.

What problems is Tonic.ai solving and how is that benefiting you?

Tonic lets us provide near-production data to our developers, which shortens cycle time for diagnosing technical issues while keeping PII away from developers outside of exceptional circumstances. Tonic's API means we can automate it, so we can invoke a sanitization run from Slack, from a scheduled Lambda, etc. It turns it into a set-and-forget tool that only needs to be touched when the database schema changes.

What do you like best about Tonic.ai?

Tonic's customer service is outstanding. They always want to ensure we are taken care of, that the product works as expected, and a lot more. The product's design itself is setup to be trusted. When we de-identify data, we need to know in great detail what's going on. Their design is intuitive and customizable to ensure all data we consider to be important is addressed properly.

What do you dislike about Tonic.ai?

The biggest dislike we have is that the product's data de-identification job can't be put on a cron from inside the product itself. We had to make a Lambda in AWS to call the Tonic Structural API to trigger a job.

What problems is Tonic.ai solving and how is that benefiting you?

We needed a way to take real data and de-identify it at scale. Using grep, sed, awk, regex on the command line to de-identify data isn't sustainable and a lot of unacceptable mistakes will be made. Using Tonic Structural solved that problem.

What do you like best about Tonic.ai?

The best part I like about Tonic is that it's very easy to create fake data. If we see in earlier scenarios, it is cumbersome to create test data as we might manually use some combination or need to write a python script to create random data. This testdata might not cover all possible scenarios actually happening in the production environment and we might miss key observations.

The data generated by Tonic is very similar to the actual data and therefore covers up all possible scenarios. Also, Tonic's support team is very helpful where in if you face any issues, you can chat with them, They are quick to respond and help you out

What do you dislike about Tonic.ai?

One thing which in my opinion is difficult about Tonic is lack of sufficient documentation and videos on how to use it.

As a newcomer, if I am using Tonic for the first time, then I need to do quite a research to understand how to use the tool especially the advanced topics like AI Synthesizer, complex generators

A detailed documenation on how to use (considering even the QA community having not known most of the Data Scienc concepts) or even videos would be really helpful

What problems is Tonic.ai solving and how is that benefiting you?

It is helping us in creating fake data easily. Fake data that is very similar to production data

What do you like best about Tonic.ai?

Prior to Tonic, nothing in our demo environment was functional. A lack of quality demo data prevented employees and partners from training and onboarding with the full functionality of our platform. We found Tonic to be the only technology that met our needs for realistic demo data generated based on production at database scale. In particular, we appreciate its support for multiple databases, automated schema change detection, and easy-to-use subsetting.

Now, with the realistic data we’re getting from Tonic, most of our reps come out of training saying, ‘I’m more comfortable with onboarding calls and demoing the platform, than with discovery calls.’ It’s having a massive impact.

Data that used to take days or weeks to manually source is now refreshed and provisioned overnight, with generation jobs completed in a matter of hours. For smaller subsets, jobs complete in as little as 30 minutes. The ease of getting this data and the quality and realism of the scenarios it captures transformed our broken demo environment into a platform driving success across the company.

You can’t tell that our demo environment runs on Tonic data. It is so close to a mirrored experience for what our partners deal with, and that helps us empower them and guide them better.

What do you dislike about Tonic.ai?

The team can be a little salesy/pushy at times

What problems is Tonic.ai solving and how is that benefiting you?

Providing a sandbox environment for our sales reps to explore and practice with.

What do you like best about Tonic.ai?

Anyone who is faced with the challenge of needing to move quickly and to be laser focused on their mission to provide value for their customers, go with the cloud hosted version of Tonic. It’s the best way to support that mission more broadly for your organization and to make sure that you're able to leverage the time and talent of your team in a way that is in support of that goal.

What do you dislike about Tonic.ai?

Some additional automation (like scheduled data generation jobs) would be helpful, but their API made it easy to build this ourselves.

What problems is Tonic.ai solving and how is that benefiting you?

As a company that's dealing with a lot of PII data, there are a lot of complex scenarios that we can't recreate manually. If I think about what it would cost for us to build something even remotely viable for us to solve our test data problem in the way that Tonic has solved it for us, it's orders of magnitude more than what it costs us to run Tonic Cloud.