Transposit automates workflow with a human touch – The New Stack


For Tina Huang, the idea of ​​automation without human intervention is misguided. This comes from his studies at the University of Chicago on cultural anthropology and blogging, as well as his stints at Google and Twitter.

“I worked at Google News and then later moved to Twitter. And I had this passion around the human in the loop, right? – like how can you augment artificial intelligence with a human editorial?” said Huang, found and chief technology officer at Transpose.

The San Francisco-based startup focuses on integrating tools used in DevOps and site reliability engineering (SRE). It aims to provide a single source of truth to calm chaos through automation and human action.

She argues that the majority of workflow tools, like Zapier and others aspire to full automation without human intervention.

“It’s really good,” she said, “but it’s very, very limited in the category of issues that can help.”

People often suggest that she will be out of a job if people can fully automate everything. His answer :

“Well, that’s true if you have a system in a silo that has no changes internally and no changes externally.”

An incident may occur with the change: I add new features to my codebase. I’m adding a new framework to my stack. But changes can also occur from the outside.

“Like on Twitter, a lot of incidents happen because you have a different traffic pattern because of the Super Bowl or something you weren’t expecting.

“I remember the Arab Spring caused problems for Twitter because it was like a spike in usage and a different style of usage than we expected,” she said.

Modeled on SQL

Huang and Adam Leventhal founded the company in 2016, attracting Divanny Lamas of Splunk as CEO. Leventhal and Lamas return to Sutter Hill Ventures, one of its financial partners. The company raised $50.4m in total after a $35m Series B round a year ago led by Altimeter Capital Management.

Huang saw the world change for developers, rather than building something from scratch, it was becoming increasingly common to piece together components from various data sources connected by APIs.

They spent three years building a platform to bring all the workflow tools together in one place. In their Series A presentations, they described it as a relational API database.

“Imagine being able to just query your data with the ease of something like SQL, and not have to worry about whether that data is coming from GitHub, Slack, any of these different sources?” Huang said.

Looking at the technology, they realized that DevOps and IT operations were most likely to respond to the idea of ​​human response in the loop, even directing machine learning to the tasks at hand.

They modeled their technology on that of SQL, which gave developers a way to talk about the data they were trying to retrieve without having to think about how to actually access the data:

“At the very heart, this is what Transposit’s integration technology is based on. So you could write an SQL query, and we’d take care of the authentication, paging, some kind of query throttling, etc.…

“One of those areas that sort of differentiates us from pretty much every integration platform I’ve seen is how deep we’ve kind of pushed this to where it’s not just a thin wrapper in addition to API calls, but it really frees the developer from thinking about the underlying services.

Automated documentation

They first applied their ideas to incident management runbooks, that workbook some companies create for those 2 a.m. calls when something goes wrong. On these occasions, the duty engineer may sweat the details of code he didn’t write or operating procedures he doesn’t know.

Transposit’s platform allows organizations to create automated runbooks showcasing a set of DevOps processes to run across multiple tools.

Huang admits that runbooks aren’t a sexy topic to talk about, but argues that’s largely because people have been pretty uncreative with them – they tend to be flat wiki pages that everyone knows, or at least assume, are not up to date, so they don’t use them.

Transposit’s answer to this to make them more interactive and useful by aggregating semi-structured and unstructured machine data with human data.

You can set a trigger, which starts an automated workflow on how to proceed. This can include things like starting a Slack channel, filing a ticket for it, assigning a commander, notifying customer service that this incident is in progress, determining which team is involved, and calling the call person.

“It can be anything from flat documentation to adding buttons that help you, so you can have a button that says, ‘restart service’, ‘update status page’, etc. .” Huang said.

It can break down silos and unify workflows between development and operations with hundreds of prebuilt integrations or easily connect to any API. Then you can also have a cleansing process.

At the same time, it documents every step of the process, providing a full audit trail and post-mortem report. Runbooks are not static documents. The underlying machine learning algorithms can examine how engineers recovered and provide reviews based on process data. It can also capture institutional knowledge.

In one blog post Explaining the need for codified processes, both machine and human, Huang wrote:

“The ideal world is being able to tie these two pieces together seamlessly. When you allow humans to put pieces of automation together, you actually end up with more automation and more robust automation.

“It’s because you can let human intuition step in where the automation logic would end up being so complex that it would cause more problems than it solved, where the logic would be so tightly tied to the specifics of a rapidly changing product it would need to be updated almost as frequently as the product itself.

Tickets Plus Action

Late last year, Transponsit announced Activitiessupporting ticketing systems like JIRA and ServiceNow.

The company doesn’t call them tickets because it usually requires action beyond the ticket itself. There’s usually an accompanying document, a Confluence page perhaps, that tells you how you’re supposed to use that JIRA issue – categorize the incident by those circumstances, for example. At the same time, people are talking about the problem on Slack or Microsoft Teams. There are repeatable actions required to file any ticket – you may need to complete these 10 fields.

“One of the nice things about the DevOps role is that a lot of it can be automated as well,” Huang said. “And so some parts of them that require human intervention like this human router for shipping – we hear a lot of people still using some form of NOC (network operations center). … honestly, we’ve even heard of people outsourcing them to third parties. Their job is literally to look at some charts and figure out which team should actually paginate. To the right? So there is a slight human element.

“But once a human identifies the team to contact, the rest could be automated.”

No-code builder

The company presents Transposit as a no-code platform.

In one BetaNews article on the 2022 forecast, Ed Sawma, vice president of operations at Transposit, postulated:

Low-code and no-code technologies have opened up new avenues for innovation by allowing more users within an organization, with or without technical experience, to contribute and add value to projects or to workflows. In 2022, we will see DevOps practices being more widely adopted in all functions outside of traditional software development. In addition to the growing adoption of no-code tools, non-technical users will experience all the benefits of a continuous delivery approach and will want to adopt it for everything they create.

The company has a no-code platform as well as a more advanced development platform. Its no-code constructor can be associated with git repositories. It is written in Python and designed to be very developer-friendly.

“What we have seen from our first customers is [that] they’re thrilled that they don’t have to wait for us to build something for them. You don’t even have to wait for us to build you a data connector,” Huang said.

“We are built on open standards with OpenAPI. So you know, if you have an internal system, anything that’s an API, you can connect to our system. … You can build almost anything you want. You know, we’ve seen customers chain it… transpose it to trigger AWS Lambda to invoke edge nodes in their compute clusters. And so, it’s very powerful and very, very flexible.


About Author

Comments are closed.