Sleuth aims to utilize Artificial Intelligence (AI) to assess the productivity of software developers | Technology

Sleuth aims to utilize Artificial Intelligence (AI) to assess the productivity of software developers | Technology
Sleuth aims to utilize Artificial Intelligence (AI) to assess the productivity of software developers 

Sleuth plans to utilize Artificial Intelligence to assess software developer productivity

During the epidemic, CEOs voiced fear that productivity would suffer as knowledge employees, particularly software engineers, switched to remote work. Although the data is varied, remote work worsened many of the issues that employees already faced in the software business. Slow feedback loops throughout the software development process were ranked second only to poor communication between teams and functional groups in a 2021 Garden poll, with the majority of developers citing them as a source of irritation. Seventy-five percent stated time spent on certain chores is wasted, implying that it might be better spent.

READ MORE: Snapchat's Newest Feature: Virtual Reality Shopping | Technology

Sleuth, a tool that interfaces with current software development toolchains to give insights to assess efficiency, was established by three former Atlassian workers, Dylan Etkin, Michael Knighten, and Don Brown, in pursuit of a solution to boost developer productivity. Sleuth reported today that it has secured $22 million in Series A investment led by Felicis with participation from Menlo Ventures as well as CRV. According to CEO Etkin, the funds will be used to build new products and grow the company's employees (specifically the engineering and sales teams).

READ MORE: 7 Incredible Mixed Reality Solutions to Review in 2022 | Technology

"With the pandemic's avalanche of remote activity, the requirement for developers, managers, as well as executives to understand and communicate about engineering productivity has expanded dramatically," Etkin told TechCrunch via email. "Developers who are no longer in the same room need a mechanism to collaborate around deployments and a quick way to figure out when one has gone awry. Managers want an inconspicuous approach to learning about bottlenecks affecting their teams in advance. Executives require a non-intrusive method of assessing the impact of their company's activities and investments. Sleuth removes the burden of understanding and conveying technical efficiency off-line and puts it in the hands of everyone."

READ MORE: Virtual Reality helps police officers in dealing with difficult situations | Technology

Etkin, Knighten, & Brown were Atlassian coworkers who claim to have assisted the company's engineering groups in moving from delivering software every nine months to daily releases. Before becoming the development manager at Bitbucket and StatusPage, Etkin worked as an architect on the Jira team, while Knighten and Brown were VP of product and architect/team lead, respectively.

During Etkin's tenure at Atlassian, which expanded from 50 to over 5,000 workers during the period Sleuth's co-founders were there, it became "crystal evident" that many engineering teams lack a quantitative approach to gauging efficiency — and that this gap may prevent them from expanding and developing.

READ MORE: Using Virtual Reality to Provide the Benefits of Learning | Technology

"Measuring engineering performance is a well-known, big, and rising challenge that has recently been solved. "The requirement for insight into engineering productivity has increased as every organization invests more substantially in software engineering," Etkin added. "However, evaluating efficiency has traditionally been difficult due to a variety of factors, including technological complexity, a lack of data, and the use of dubious proxy measures that encouraged micromanagement and distrust."

Sleuth's solution is DevOps Research as well as Assessment (DORA) metrics, a growing standard used by development teams to track things like how long it takes to deploy code, how long it takes for a service to recover from failures, and how often a team's improvements result in difficulties after deployment. DORA developed from a Google academic research team that surveyed over 31,000 engineers on DevOps methods between 2013 and 2017 to discover the fundamental differentiators between "poor performers" and "elite performers."

READ MORE: The Dividing of the Metaverse: Meta against Varzo and Nvidia | Technology

Sleuth isn't the only platform that measures productivity with DORA metrics. The DORA standard has been adopted by alternative systems such as linear, Jellyfish, and Athenian. However, Etkin maintains that its rivals don't measure these indicators "completely or properly."

"Sleuth is different... because we utilize deployment tracking to simulate how developers ship their work from idea to launch," he said. "By accurately modeling how engineers ship throughout their pre-production as well as production environments, as well as how they connect with issue trackers, CI/CD, error trackers, as well as metrics, Sleuth can create a completely automated... perspective of a team's DORA metrics & engineering efficiency."

READ MORE: The Chief Scientist of Reality Labs Describes a New Compute Architecture for True AR Glasses | Technology

From current systems like Datadog and Sentry, Sleuth attempts to figure out a team's baseline change failure rate (i.e. the percentage of modifications that resulted in degraded services) and mean time to recovery - two of the four DORA metrics. According to Etkin, the platform can detect when a measure deviates from the baseline and even automate processes in the development process to potentially enhance the statistic.

Individual teams can track their DORA stats through Sleuth's project dashboard. Trends between projects and teams are revealed through an organization-wide dashboard.

READ MORE: The Top Priorities for Artificial Intelligence in Banking in 2022 | Technology

"Customers point Sleuth to... error data, and Sleuth warns engineers when these metrics exceed the failure threshold," says the company. Engineers may be able to concentrate on their work rather than needing to understand every statistic in their system or what 'normal' looks like for each measure if AI is used to determine these values."

Of course, DORA measurements aren't everything. When a company's attention on them becomes all-consuming, they might become a barrier. "Developer productivity should not be judged by the number of defects, delayed deliveries, or incidents," said Sagar Bhujbal, vice president of technology at Macmillan Learning. It creates unnecessary anxiety among development teams, which are constantly under pressure to produce more capabilities quicker and better."

READ MORE: Why Hybrid Intelligence is the Future of Artificial Intelligence at McKinsey | Technology

Etkin agrees, highlighting the necessity for technical managers to avoid micromanaging.

"Engineering is a creative effort, and engineers resemble artists more than assembly-line employees," Etkin added. "Supervise the proper metrics, precisely measure them, as well as offer engineers with the tools they need to improve the metrics," engineering managers should say.

READ MORE: How fintech firms are utilising AI and machine learning to generate alternative loan scores | Technology

LaunchDarkly, Puma, Matillion, and Monte Carlo are among Sleuth's customers, who range from large corporations like Atlassian to small businesses like LaunchDarkly, Puma, Matillion, and Monte Carlo. According to Etkin, the platform has monitored about a million deployments and performed over a million automated activities on developers' behalf. When questioned about sales figures, he replied that Sleuth, which employs 12 people, grew 700 percent last year and has a "very strong" margin and cash flow.

Source: Kyle Wiggers, Tech Crunch, Direct News 99