Topics
Latest
AI
Amazon
Image Credits:Thomas Barwick(opens in a new window)/ Getty Images
Apps
Biotech & Health
Climate
Image Credits:Thomas Barwick(opens in a new window)/ Getty Images
Cloud Computing
Commerce
Crypto
Image Credits:Glean
Enterprise
EVs
Fintech
Image Credits:Glean
fund raise
gizmo
punt
Government & Policy
Hardware
Layoffs
Media & Entertainment
Meta
Microsoft
privateness
Robotics
Security
Social
infinite
startup
TikTok
transit
Venture
More from TechCrunch
event
Startup Battlefield
StrictlyVC
Podcasts
Videos
Partner Content
TechCrunch Brand Studio
Crunchboard
Contact Us
GenAI has its emergence . But if there ’s one thing it excels at , it ’s surfacing solution from vast pools of data point .
EnterGlean , whose software connect to go-ahead first- and third - party databases to field unmistakable - English petition ( for instance “ How do I invest in our company ’s 401k ? ” ) from employee , sort of like a customChatGPT . Launched by Arvind Jain , the Colorado - laminitis of cloud data management company Rubrik , Glean was inhale by Jain ’s observance that Rubrik employee often skin to notice the information they needed to do their job — and that staff member at other companies were struggling with the same .
“ I check that engineers were spending too much time outside code , account manager could n’t regain the latest research or presentation needed to close spate , new employees took too long to onboard , and so on ” Jain assure TechCrunch in an interview . “ This growing trouble destroy productiveness , run down vigor and detracted from the employee experience . ”
It seems Jain was on to something .
A recent Gartnersurveyfound that 47 % of desk actor have bother line up the data they demand to do their jobs . In the same sketch , workers reported that the grow number of apps they have to manage at work — 11 on mediocre now versus six five years ago — is exasperate the challenge .
In 2019 , Jain — along with a small founding squad — establish Glean as an AI - power search app pitch toward enterprise customers .
The first few iterations were along the line of Microsoft ’s SharePoint Syntex and AmazonKendra , occupying a production class known as “ cognitive lookup . ” Using natural lyric processing , the early Glean could sympathize text file minutia in increase to searches employees across an organization might execute .
Join us at TechCrunch Sessions: AI
Exhibit at TechCrunch Sessions: AI
Over the year , Glean evolved into a platform that connects with and analyzes a company ’s databases and information storage to respond employee inquiry — following after the explosive GenAI trend . Glean today consume information from source let in livelihood tickets , chat messages and customer relationship management platform entries and applies GenAI to attempt to turn that all that into insights and relevant answers .
One imago companies would be wary of connecting their proprietary datum — especially their inner schmoose data — to a GenAI platform that do this deep a storey of scraping and analytic thinking . And that would n’t be an incorrect supposition .
A late Ciscopollfound that more than one in four organisation have banned the exercise of GenAI over privacy and data security risks . In the poll , society said they fear GenAI tools would compromise their IP or potentially disclose other raw information to the world — or their competition .
But Jain assert that Glean is “ strong ” and “ private ” — at least to the extent a swarm - based GenAI platformcanbe .
“ Glean respects the same permission jell in a caller ’s data root ( Slack , Teams , Jira , ServiceNow , etc . ) , so employee only welcome solution based on the data they ’re allowed to access , ” Jain said . “ When a user deletes a written document in the underlie app , the document gets deleted from the Glean system . ”
What about the curse from which most GenAI suffers , though — hallucinations ? Is Glean resistant from making up fact and cite , getting summaries wrong and missing the dot of canonic request ?
It ’s possible ; this writer was n’t able-bodied to test Glean himself . But Jain , while refusing to say just how often Glean hallucinates , highlight the mitigations in place to make the weapons platform ’s GenAI more reliable , including a simulation trained on customer data point to learn industry and fast - specific jargon and let customer switch among several open informant GenAI models to drive Glean ’s effect experience .
“ AI workplace help need to render personalised issue establish on who ’s searching , ” Jain state . “ Various aspect of the searcher — their purpose , job subroutine , management hierarchy , specific projects and responsibilities and even who they make with — terminate up being important in limit the cognitive content that ’s relevant to them . Glean learn a custom good example for every customer to deliver highly personalized results to every employee based on these attribute . ”
Glean also employs RAG ( forgetful for Retrieval - Augmented Generation ) , an progressively coarse technique used to “ grind ” GenAI by retrieving information from outside source of noesis , to boost performance . Jain sound out that every answer Glean pay is “ fully referenceable ” back to the original informant .
“ Glean [ can urge the ] documents drug user might need for their day - to - day work by learning from past work patterns , ” Jain tell . “ [ It ] delivers turnkey execution of a complex AI ‘ ecosystem , ’ with over 100 connectors . ”
Glean makes money by charge a monthly per - seat subscription , based on annual contract .
Despite competition from vendors like Microsoft ( specifically Copilot ) and OpenAI ( ChatGPT ) as well as endeavour search supplier such as Coveo , Sinequa and Lucidworks , Jain say that job has been quite unassailable as of late , with annual recurring revenue tight to quadrupling in the last class .
That runs counter to the story that corporations — far from embracing GenAI wholeheartedly — have been slow and wary to deploy it across their line occasion .
Responding to a December 2023surveyby Convrg.io , the Intel subsidiary , only 10 % of governance say that they ’d launched GenAI solutions all the path to production in 2023 . The huge absolute majority of root stay in the research and examination stage , the organizations enounce — imply companies have n’t been successful in find money - making GenAI use cases .
Glean ’s financials — and a 200 - strong client base that let in Duolingo , Grammarly and Sony — appear to have won over investor , however .
Glean today foretell that it raised $ 200 million in a Series D funding round cobalt - direct by Kleiner Perkins and Lightspeed Venture Partners with involution from General Catalyst , Sequoia Capital , Adams Street , Coatue , ICONIQ , IVP , Latitude Capital and extra strategic backers Capital One Ventures , Citi Ventures , Databricks Ventures and Workday Ventures .
Kleiner Perkins ’ Mamoon Hamid had this to say in a command : “ The opportunity for Glean is enormous , and we have so much sentence in the squad ’s ability to ply the GenAI root for the endeavor that we co - led this round after investing in every round prior to this , after result their Series A in 2019 . I ’ve expend my venture life history investing in applications that enable knowledge workers to be more productive , whether it ’s Slack , Box or Figma , and see huge potential in Glean to exchange the direction that people wreak . ”
Jain suppose that the unexampled capital , which brings Glean ’s total bring up to ~$360 million and values the inauguration at $ 2.2 billion , will be put toward expand “ all of ” Glean ’s teams ( the Palo Alto - based company has ~300 employees at present tense ) , enhancing its product and “ build out a robust go - to - market motion . ”
“ Glean has continue to see strong and growing client requirement , peculiarly from enterprise who ’ve spent the past yr evaluate the necessary requirements to take GenAI into their organization , ” Jain said . “ We ’ve always been prudent in hiring and outgo , and the late gain in lease is to meet strong customer demand . ”
