Experimenting with Bots

2016 was a significant year for bots, artificial intelligence, and anything around machine / deep-learning technologies and tooling. Some might say that it was the year that a lot of this came into view for the average consumer because of products like Google Home and Amazon’s Echo and Alexa.

Regardless of your position on bots for the short and/or long-term, I believe that we all can agree that their potential to radically influence the way we perceive and engage with the world is, at the very least, significantly large.

With that in mind I decided (with a bit of a challenge from @Jeff) to explore what it might look like to have some bot-action with the data that we’re currently spitting out in our current Alpha Release.

So, being familiar with building Slackbots (via a previous project / venture) I decided to start there and in a day I had built out some core concepts and some prototype(s) of what it might look like.

Bots, everywhere!

I thought I’d share a few of these with you here (via gifs)!

And, for some strange reason “Sir Pinpoint” (which is what I lovingly call him – a temporary name, of course) is spitting out duplicate responses at the time of filming, so, I’ll have to take a look at why that’s happening post-publication. But, the gist is there. With that, take it away @Sir_Pinpoint!

More info on the team

One of the things that Slack has done within their latest updates of their API is the ability to create deeper interactivity with the bot through buttons (and even the coloration of those buttons).

What we can do with our existing data is ask @Sir_Pinpoint for information on the given software team and provide visual cues as to how folks are doing. Perhaps, for instance, anything person in “Red” might be flagged for deeper review or this person might need help with their current work or sprint or an issue / bug. Who knows!

But, the possibility of having access to team dashboards / metrics at your fingertips is exciting. Managing a team of software developers is already tough enough – with Pinpoint our goal is to help make those challenges much easier and it starts with accessibility of data where you are already working (i.e. in Slack).

Old issues?

So, what about asking @Sir_Pinpoint about some of the oldest issues in a repo? Why, yes, why not?!

As you can imagine, having this information available to you immediately and on-demand is powerful. This is just a taste of what is possible and if these prototypes showcase anything, it could be a very delectable meal.

What about language distros?

Showing graphics of the data that we’re mining and reviewing of your current organization, projects, and teams could just be the start. Obviously, @Sir_Pinpoint could summon up a treasure trove of visualizations of your data, if connected properly.

Most contributions in last 7 days?

Allowing for more fine-tuned queries, such as date-based queries, could be really neat as well. We already calculate and display 7, 30, 60, 90 days in our dashboarding right now but you could technically ask for any window of time and even ask for specific date ranges.

Oh, the possibilities!

Contributions by activity?

And what about contributions by activity? Other visualizations! The gif got cut-off on the output so that’s below:

“Contributions by Activity” provided by @Sir_Pinpoint

Active repos in 60 days?

Finally, what about the most active repos in the last 60 days? Well, there you go. Output is also below:

“Active Repos in the last 60 days” provided by @Sir_Pinpoint

Of course, this is just a few of the prototype outputs that I programmed in @Sir_Pinpoint but you can get the gist and you can begin to see the value that bots could bring when done well (and this is just the start).

This is part of the much broader and bigger vision we have around #EngineeringOps (or #EngOps for short) and we think that giving developers, technical managers, and software teams faster access to the data that they need is of high importance.

Oh, and that’s why we started building Pinpoint to begin with!

(Note: We are not currently building any production-ready bots at this stage! But, we could use your feedback on our Alpha! Get in touch!)