Get ready to find out how the Augoor’s Productivity Hack Challenge led to these conclusions.
Artificial intelligence constantly impacts billions of people’s lives worldwide, augmenting work experiences for professionals in different fields. At Augoor, we decided to leverage the power of machine learning algorithms, deep learning (neural networks), and natural language processing to build a Code Intelligence platform that lets devs quickly explore, understand, and document code. We’re like a navigation tool but for devs.
Quickly navigating through big-cluttered codebases, finding the right code snippets, and having them automatically documented on the spot sounds like a big time-saver on paper. But we wanted to capture Augoor’s impact on the field, so we designed a challenge to get some answers.
And this was how the Productivity Hack Challenge was born. Our research team selected a group of developers and conducted an exercise to measure their speed in solving a coding task. This challenge helped us discover that Augoor increases devs’ speed by 60% while positively impacting their coding experience.
The mission: multiplicate the output of code that dev teams generate
Did you know that devs spend at least 70% of their time reading and understanding code instead of writing it? How much more productive would they be if we could invert that percentage? Augoor’s mission is to optimize companies’ go-to-market helping dev teams multiplicate the output of code they generate.
Moreover, we allow devs to stay on the creative side of coding, spending more time creating solutions that innovate our world than focusing on reading old code. Also, if you are a manager or tech leader, Augoor can help you monitor, supervise, and find opportunities within multiple source codes in your organization. All in one single tool.
The challenge: a study to evaluate Augoor’s effectiveness
To perform a research activity on productivity, we had to understand it. Measuring it is no walk in the park, as the industry still has more questions than answers. We found that we should shift away from output-driven metrics and focus on ‘outcome-driven’ ones for more accurate results. We kept in mind that the dev team’s perceptions of well-being, satisfaction, and productivity are intricately connected.
That’s why we decided to inspire our productivity research in a holistic approach, using the SPACE framework to choose which aspects to investigate. We selected activities and questions to help us measure this framework’s efficiency and flow dimension. Some of the metrics we focused on were:
- Time measured through a system: coding tasks resolution time.
- Perceived ability to stay in flow and complete work:
- Were you able to filter search results faster thanks to the AI-generated explanations?
- Did you find code to reuse faster, thanks to our tool?
- Did you read less code thanks to the AI-generated explanations?
We selected groups of developers working on the same projects and challenged them to solve some of their daily coding tasks with and without Augoor under the same conditions. Like in clinical trials, we measured the product’s impact by comparing each scenario’s outcomes and asked participants to complete a survey about their experience using Augoor.
Findings: Augoor augments devs’ speed and satisfaction levels
Let’s take a deeper look at the outcomes and the feedback we gathered from the challenge participants.
#1: Our Code Intelligence platform speeds up the resolution of coding tasks
As mentioned before, our probing activity was based on the efficiency and flow aspect of the SPACE methodology, which includes individual, team, or system metrics that measure improvements in activities that users often refer to as software development waste. Finding a specific function, snippet, or file could mean hours wasted. So, at a system level, we decided to measure how much faster devs would complete a code retrieval task using Augoor, and the result was an increase of 60% in speed.
“Without Augoor, I’d have sent several emails to find the code I needed and waited for it to arrive for days. Today, searching with Augoor, I identified the code I was looking for in the first search result, how it is being used and who is doing it.” Engelbert Quiroz, Tech Lead
#2: Augoor positively impacts devs’ satisfaction
A happy developer can spend more time on creative tasks and less time trying to find, read, and understand other people’s code. These are some of the insights we got from the participant’s survey answers:
- 100% believed that they found the code they were looking for and the code to reuse faster, thanks to Augoor.
- 94% reported reading less code to solve the proposed task, thanks to Augoor.
- 71% thought that Augoor would speed up onboarding.
- 59% believed that Augoor helped them understand code faster.
“I find the tool useful for searching for code, not ‘reinventing’ the wheel, and looking at other ways to solve a problem I may have.” Iván Florez, Java developer
“Augoor is a good tool, especially for searching in unknown repositories, and could greatly help the adaptation of new members to work teams.” Daniel Ospina, Java developer.
The Augoor Productivity Hack Challenge is an ongoing initiative meant to be scaled to continue measuring the platform’s impact on productivity. As we introduce its benefits to more teams, we’ll continue to challenge more devs and listen to their feedback. This way, we’ll get a recurrent perception of Augoor’s role in smartening up software development.