If you’ve ever shopped online, chances are you’ve encountered a popup asking something along the lines of, “Hi, how can I assist you today?” Thanks to advancements in artificial intelligence, chatbots have become an important way for businesses to conduct customer service and online marketing operations. Chatbots have grown somewhat ubiquitous: Gartner predicts that 85% of customer interactions will be with some form of AI application.
Programmers design chatbots with all sorts of levels of sophistication, and perform a wide variety of tasks. They’re especially well-suited for mobile users communicating on messaging platforms. Businesses have the option of buying a third-party bot to deploy within their digital presence, or choose to develop their own in-house.
Chatbots have also begun to impact the talent recruitment space. Here’s a primer on what chatbots can do for recruiting, and how the technology connects with developer skillsets.
What goes into modern chatbot development?
It starts with picking a platform—a messaging service like Messenger, Slack, Discord, or an internal system which could be built on any variety of ERP platforms—Oracle, SAP, Microsoft Dynamics, AWS, and the like. A business can build a bot from scratch, or there are plenty of coding frameworks specifically for chatbots. Some prominent choices include: IBM Watson, Microsoft Bot Framework, LUIS, Wit.ai, Api.ai, and Chatfuel.
The workflow for creating a chatbot follows a similar path as website and application development. It should be thought of in four phases: Design, Building, Analytics, and Maintenance. Qualified engineers will have experience managing end-to-end software development lifecycles. DevOps, Agile, and Scrum workflow methodologies apply in developing chatbots.
How chatbots are impacting the recruitment space
Talent recruiters are implementing chatbot technologies into their processes. A recruitment chatbot often helps with low-level administrative tasks on the recruiter’s end, and make candidates’ experience more targeted and positive. Some examples:
- According to CareerBuilder, 67% of job seekers prefer consistent updates during the hiring process. A different poll by Workopolis states 43% of candidates never receive a response from a job application. There’s opportunity here for chatbots to orchestrate communications in this situation and improve the experience.
- The application process can be lengthy and difficult. One CareerBuilder survey found that 60% of job seekers abandoned applications in the middle of the process. A chatbot potentially can assist applicants who might get “stuck” while filling out the required information.
- 77% of survey respondents told the Undercover Recruiter that job descriptions are too vague and need more information. A chatbot that fields applicant questions poses a viable remedy in this situation.
- A chatbot is an effective way to ask basic screener questions in lieu of a time-consuming and often untargeted data capture. This offers a way to obtain better information from job candidates, and lighten the load on applicants.
Common roles in chatbot development
If you’re recruiting into the chatbot development space, it helps to understand how programmer roles fit into the development cycle.
- A user experience design analyst participates in design stages of production. This often means preparing chatbot storyboards, process flows, and sitemaps that guide how the chatbot interacts with a live user. During testing phases, this role conducts user research and evaluates feedback in making improvements and fine-tuning the UX for the bot application. A strong UI/UX design background or experience in front end development usually is preferred here.
- An applications systems engineer builds and deploys a chatbot application. This role is hands-on with software, coding, and frameworks involved in chatbot development. In machine learning, teams might use something like Splunk to capture machine data in a searchable repository. Hibernate ORM is an object-relational tool for Java programming used by machine learning developers, used in conjunction with a relational database language like Oracle-proprietary PL/SQL, Microsoft SQL Server, or the open-source MySQL. The open-source programming language Python—particularly the Anaconda distribution—is one of the most-used languages in machine learning.
- A Java full stack engineer or a similar variant typically reviews and modifies chatbot development, checking coding and constructions throughout the application—frontend middle-tier, persistence, and backend. This role collaborates with the business side of the development company and ensures development of the product will meet business objectives. Sometimes responsibilities include creating custom system-to-system interfaces to aid in production.
