Siadhal Magos and Shahriar Tajbakhsh were working at Uber and Palantir, respectively, once they each realized that hiring—particularly the interview process—was becoming increasingly cumbersome for a lot of firms' human resources departments.
“We realized that crucial a part of the hiring process is the interviews, but in addition probably the most opaque and unreliable part,” Magos told TechCrunch. “Additionally, taking notes and providing feedback requires a variety of effort that many interviewers and hiring managers attempt to avoid in any respect costs.”
Magos and Tajbakhsh felt the hiring process was ripe for disruption, but desired to avoid abstracting an excessive amount of from the human element. So they began Meta viewan AI-powered note-taking app for recruiters and hiring managers that records, analyzes and summarizes interviews.
“Metaview is an AI notator designed specifically for the hiring process,” Magos said. “It helps recruiters and hiring managers focus more on attending to know candidates and fewer on extracting data from the conversations. This saves recruiters and hiring managers a variety of time writing notes and helps them be more present during interviews because they don’t should multitask.”
Metaview integrates with apps, phone systems, video conferencing platforms, and tools like Calendly and GoodTime to routinely capture interview content. Magos says the platform “takes into consideration the nuances of recruiting conversations” and “enriches itself with data from other sources,” reminiscent of applicant tracking systems, to spotlight probably the most relevant moments.
“Zoom, Microsoft Teams and Google Meet all have built-in transcription, providing a possible alternative to Metaview,” Magos said. “But the knowledge Metaview’s AI extracts from interviews is much more relevant to the recruiting use case than generic alternatives, and we’re also helping users take the subsequent steps of their recruiting workflows in and around these interviews.”
Photo credit: Meta view
There are actually things that go improper with traditional interviews, and a note-taking and conversation evaluation app like Metaview could, at the very least in theory, help. As an article in Psychology Today notes: The human brain is filled with prejudices that hinder our judgment and decision-making, for instance the tendency to rely too heavily on the primary piece of knowledge presented and to interpret information in a way that confirms our pre-existing beliefs.
The query is: Does Metaview work – and more importantly, does Metaview work equally well for all users?
Even one of the best AI-powered voice dictation systems suffer from their very own biases. A Stanford study showed that the error rate for black speakers on speech-to-text services from Amazon, Apple, Google, IBM and Microsoft is high almost twice as high as white speakers. Another, more moderen study published within the journal Computer Speech and Language found statistical significance Differences in the best way two leading speech recognition models treated speakers different genders, ages and accents.
There are also hallucinations that have to be taken into consideration. AI makes mistakes when summarizingalso in Meeting summaries. In a recent article, the Wall Street Journal cited a case by which an early adopter using Microsoft's AI Copilot meeting summarization tool Copilot invented participants and implicit calls revolved around topics that were never discussed.
When asked what steps, if any, Metaview has taken to mitigate bias and other algorithmic issues, Magos claimed that Metaview's training data is diverse enough to supply models that “exceed human performance” in recruiting flows and customary benchmarks for Bias do well.
I'm skeptical and somewhat suspicious of Metaview's approach to the way it handles voice data. According to Magos, Metaview stores conversation data for 2 years by default unless users request the information be deleted. That looks as if an exceptionally very long time, and the candidates probably would too.
But none of this appears to have affected Metaview's ability to draw funding or customers.
Metaview raised $7 million this month from investors including Plural, Coelius Capital and Vertex Ventures, bringing the London-based startup's total raised to $14 million. Metaview's customer base totals 500 firms, including Brex, Quora, Pleo and Improbable, in response to Magos – and has grown 2,000% year-over-year.
“The money will primarily be used to expand the product and engineering team and add fuel to our sales and marketing efforts,” Magos said. “We will triple the dimensions of the product and engineering team, further refine our conversation synthesis engine in order that our AI routinely extracts exactly the correct information our customers need, and develop systems to proactively detect issues reminiscent of inconsistencies within the interview process and seemingly existing candidates. “lose interest.”
