Topics
Latest
AI
Amazon
Image Credits:lemono / Getty Images
Apps
Biotech & Health
mood
Image Credits:lemono / Getty Images
Cloud Computing
Commerce
Crypto
Image Credits:Metaview
endeavor
EVs
Fintech
fundraise
contrivance
Gaming
Government & Policy
Hardware
Layoffs
Media & Entertainment
Meta
Microsoft
privateness
Robotics
security system
societal
infinite
Startups
TikTok
Transportation
Venture
More from TechCrunch
case
Startup Battlefield
StrictlyVC
Podcasts
Videos
Partner Content
TechCrunch Brand Studio
Crunchboard
Contact Us
Siadhal Magos and Shahriar Tajbakhsh were working at Uber and Palantir , severally , when they both derive to the realization that hiring — particularly the process of interviewing — was becoming unmanageable for many corporate HR department .
“ It was clear to us that the most important part of the hiring process is the interviews , but also the most unintelligible and unreliable part , ” Magos separate TechCrunch . “ On top of this , there ’s a bunch of toil associated with taking banknote and publish up feedback that many interviewers and lease managers do everything they can to avoid . ”
Magos and Tajbakhsh think that the hiring cognitive operation was ripe for kerfuffle , but they wanted to avoid lift away too much of the human element . So they launchedMetaview , an AI - power note - take app for recruiters and hiring managers that records , canvass and summarizes problem interviews .
“ Metaview is an AI note - taker build specifically for the hiring process , ” Magos said . “ It aid recruiters and charter manager centre more on produce to lie with candidates and less on extracting data from the conversation . As a consequence , recruiter and hiring managers save up a ton of time pen up Federal Reserve note and are more present during audience because they ’re not having to multitask . ”
Metaview integrates with apps , telephone set organisation , videoconferencing platforms and peter like Calendly and GoodTime to automatically capture the content of interviews . Magos read the platform “ accounts for the nicety of recruiting conversation ” and “ enrich itself with datum from other sources , ” such as applicant tracking systems , to highlight the most relevant import .
“ Zoom , Microsoft Teams and Google Meet all have transcription built in , which is a possible choice to Metaview , ” Magos said . “ But the info that Metaview ’s AI pull out from interview is far more relevant to the recruiting use case than generic choice , and we also serve users with the next steps in their recruiting workflows in and around these conversations . ”
Certainly , there ’s plenty awry with traditional line of work interviewing , and a bank note - taking and conversation - analyzing app like Metaview could help , at least in possibility . As a piece in psychological science Today notes , the human brain is prevalent with biasesthat hinder our judgement and decision making , for object lesson a tendency to rely too hard on the first piece of information offered and to translate information in a way that confirms our preexisting belief .
Join us at TechCrunch Sessions: AI
Exhibit at TechCrunch Sessions: AI
The interrogation is , does Metaview work — and , more importantly , lick equally well for all users ?
Even the unspoilt Army Intelligence - powered speech dictation systems suffer from their own biases . A Stanford study showed that error rate for pitch-dark verbalizer on speech - to - text services from Amazon , Apple , Google , IBM and Microsoft arenearly double those for white speakers . Another , more late study published in the daybook Computer Speech and Language find statistically significantdifferences in the agency two run speech identification models treated speakersof different genders , ages and accents .
There ’s alsohallucinationto consider . AI makes mistakes summarizing , including inmeeting summaries . In a late story , The Wall Street Journal cited an illustration where , for one other adoptive parent using Microsoft ’s AI co-pilot tool for sum up meetings , Copilot invented attendees and implied call were about subjects that were never discuss .
When inquire what steps Metaview has conduct , if any , to extenuate bias and other algorithmic payoff , Magos claim that Metaview ’s training data is various enough to yield models that “ surpass human performance ” on enlisting workflows and perform well on pop benchmarks for prejudice .
I ’m sceptical and a bit wary , too , of Metaview ’s overture to how it handles delivery datum . Magos order that Metaview shop conversation data for two years by default option unless users quest that the data be delete . That seems like an exceptionally long meter .
But none of this come along to have affected Metaview ’s ability to get financing or customers .
Metaview this calendar month raise $ 7 million from investors including Plural , Coelius Capital and Vertex Ventures , bringing the London - based startup ’s total get up to $ 14 million . Metaview ’s customer enumeration stands at 500 companies , Magos say , including Brex , Quora , Pleo and Improbable — and it ’s grown 2,000 % year - over - year .
“ The money will be used to grow the mathematical product and engineering squad in the first place , and give more fuel to our sale and merchandising efforts , ” Magos said . “ We will triple the product and engineering team , further fine - line our conversation synthesis engine so our AI is mechanically extracting incisively the right information our client need and develop system to proactively notice issues like incompatibility in the consultation process and candidates that seem to be losing interest . ”
