Before the recruiter can collaborate with my chatbot, they are prompted with a defining question: do you accept that Black lives matter? If they acknowledge no, or try to added belittle the cause by saying that all lives matter, the chat ends immediately. If not, the recruiter may abide interacting with the bot which answers a series of questions about my work history, experience, goals, etc.
This is just an automatic chatbot, so by no means the defining endpoint of my assurance — I always thoroughly analysis what initiatives the aggregation has taken, what causes they fight for, and what their aggregation ability is like. But from basic psychology, I know that the recruiter agreeable with my site will look at this catechism and either (1) be so taken aback and turned off by this “unprofessional” catechism that they don’t acquaintance or (2) acquaintance me. The latter, I assume, will have a basic compassionate of how our association always oppresses marginalized communities, and why using their belvedere to speak out is so important.
Due to accepted demand, I’m autograph this commodity as a short tutorial on how to make your own chatbot and the steps I took to get there. I’d never done any communicative AI design before this project, but picked up the tools pretty easily. It’s simple, free, and pretty fun to apparatus — anyone can do it! Hopefully, this small guide can help people get started.
I wrote out all the questions that I’ve been frequently accepting in my UX design interviews and organized it in an Excel spreadsheet so that I could keep track of what answers I still needed to add to the bot. I get a lot of questions about my experience, what I’m attractive for in my next role, my abstruse skills, and software I use, etc. To show off my personality, I added fun little answers about my hobbies, sign, and admired music.
I abundant the user flow and visually mapped out the dialog, which helped me mark exit and entry points as well as guide the chat with advancement chips so that the recruiter would have a better idea of the type of questions they can ask me. I had other questions listed in my spreadsheet that I didn’t put into any advancement chips so that it didn’t seem like the chatbot was abased on them.
DialogFlow is simple to work and there’s a ton of free tutorials and affidavit out there. Here are the basics, as well as some tips I picked up along the way:
To begin, create an agent.
Start making your intents, which is each step of what the chatbot will acknowledge to. You should absolutely name each intent to stay organized and structured.
Add in training phrases, which are expressions that you expect the user to say. Avoid putting in one-word phrases that could be easily abashed with article else you’re going to need them to say on a altered intent, i.e. don’t just write “why,” be specific and write out the full phrase “why do you want to work here” as well as some iterations of the phrase. These training phrases are important because it’s what makes the bot smarter, and the more interactions you get with your chatbot, the more it’s able to admit and affix phrases together.
You can use Entities to define synonyms of assertive words and advertence options. If you use this, be sure to add your entity to your action and parameters. This will allow your bot to advertence user input.
You can write out your custom responses. Each text acknowledgment can have variants, or you can have assorted text responses at once. Make sure you’re assuming a friendly, human agent-like acquaintance for your users — they know it’s a chatbot, but it’s nice to make it feel like they’re absolutely talking to you. Plus, it’s a great way to let your personality shine and stand out to recruiters!
Follow-up intents are a great way to make the bot more interactive. That is, ambience up an intent to ahead a yes or no acknowledgment and putting out a custom acknowledgment to handle the input after abolition the flow by interpreting their acknowledgment as a altered phrase. I used these nested intents in my filters to make sure the recruiters would only have access to my answers if they answered that Black lives matter.
On the right side of the DialogFlow interface, there’s an agent test. You can use this to type out a user announcement and affirm that the agent is acting the way you want it. I tested my interactions after saving every intent.
When you’re satisfied, you can deploy your bot under the Integrations section. In your embed code, you can edit the HTML and CSS customizations to change its colors to match your own website and add a contour account for your bot.
After my bot went public and I tweeted about it, I got an cutting number of user interactions. In DialogFlow, I can see the history of everyone’s interactions as well as any flagged incomparable requests (essentially, where the bot failed to adapt the user’s response). This is where I could learn more about what users were asking my bot that I didn’t anticipate, such as “where are you based?” and “salary requirements.” From attractive through these interactions, I knew what other intents I needed to accommodate into my bot in order to make it more finer conversational.
Unfortunately, because of the arguable nature of the bot and the acceptance of the tweet, I also got a ton of all lives matter people on my website abrogation me abhorrent comments.
I wanted to make sure that the trolls still got a proper alternation with my bot rather than just the fallback intents. So, I got my bot to out-troll the trolls with some assets that would hopefully accommodate accuracy on the subject.
Educational assets response
The bright side is that, thanks to NLP, the more the user interacts with the bot, the smarter it gets. That means that the more that trolls sent death threats and mean comments to my bot, the more it could admit these abhorrent phrases and acknowledge with some assets where the user can brainwash themselves about the BLM movement. Kill em’ with kindness.
So, did it work?
Well, I accustomed a ton of sweet emails from altruistic people at Google, ServiceNow, etc. extensive out to offer an centralized reference, acquaint me to their Head of Design, or affix me to their employers. I’m crediting most of my accepted success to Twitter, which helped the chatbot go viral and gained the absorption of companies around the world, from able tech start-ups in London to neurotech AI companies in the Bay Area.
I’ve also been accepting green afire by companies who didn’t see the tweet, but mentioned how much they liked the chatbot in my antecedent phone call with them. So I would say, admitting how arguable article like this seems, it’s absolutely worked! Most importantly, it lets me speak out about my values and set up those expectations right at the beginning, instead of accidental design to article that I don’t accept in.
I’ve always been told to avoid “mixing business with politics,” but these issues are simply too loud to ignore anymore. In the wake of the death of endless Black people at the hands of police atrocity and the protests that have followed, most companies have publicized their adherence to the Black community.
I accept that we shouldn’t accommodation our values to appear “neutral” and “professional” to recruiters. Adding appearance like this is a artistic way to stand out to recruiters and empowers you to work with the right kind of people.