How 3 AI Startup CEOs Use Artificial Intelligence Tools in Their Daily Lives
AI tools are already part of the routine for people on the front lines of technology. And we’re not just talking about building products with artificial intelligence — we’re talking about actually using it, day in and day out, to solve real problems, make faster decisions, and keep up with a schedule that simply never stops.
Three CEOs of artificial intelligence startups in San Francisco opened up to Business Insider and shared how they use AI tools at work and in their personal lives. Their cases are quite different from each other, but they all share something in common: AI is no longer a promise — it has become an essential part of the workflow.
From vibe coding sessions to prototype products, to analyzing complex legal contracts and diagnosing data center operations, all the way to partnership proposals that used to take weeks and now get done in minutes — what these leaders shared reveals how technology is transforming the way people run a tech company.
Below, you’ll get to know the routines of Vipul Ved Prakash from Together AI, Arvind Jain from Glean, and Daniel Yanisse from Checkr, along with the practical ways each of them uses AI tools available today. 🚀
Vipul Ved Prakash: vibe coding, contracts, and AI-powered diagnostics
Vipul Ved Prakash, co-founder and CEO of Together AI, a platform that lets developers train generative AI models, is the kind of leader who doesn’t just stick to strategy. According to him, he uses AI for pretty much everything. The concept he practices even has its own name in Silicon Valley: vibe coding, an approach where you use language models to generate code conversationally, without having to write everything from scratch.
In practice, Vipul shared that he frequently creates prototypes and mockups of products the company wants to build, and he can do this very efficiently with vibe coding. Instead of spending weeks on specifications and requirement documents, he shows up to a meeting with something that already works, that can already be clicked and evaluated. This completely changes the decision-making dynamic within the company, because the team gets to react to something concrete rather than an abstract idea. For a startup CEO, where speed is everything, that difference is huge and represents a real competitive advantage.
But vibe coding is just one part of the story. Vipul also highlighted that AI helps him with legal contract analysis. Together AI frequently deals with complex contracts involving data centers, cloud services, and enterprise clients. Having a tool that can digest those documents and flag the most relevant points saves a considerable amount of time before even formally involving the legal team.
Another use he mentioned is quickly learning about new topics. Vipul said he spends way more time interacting with AI tools than with traditional search engines or productivity apps. When he needs to dive deep into a new subject, AI works like an on-demand tutor that can synthesize information in a contextualized way, saving hours of manual research.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
Perhaps the most impressive case Vipul shared involves Together AI’s own operations. The company manages multiple data centers and tens of thousands of GPUs. When an operational issue comes up, most diagnostics are already handled by artificial intelligence, not humans. This means AI isn’t just helping with creative or administrative tasks — it’s directly involved in the business’s critical infrastructure, identifying failures and pointing out solutions in real time. 💡
What the Together AI case reveals about productivity with AI
What stands out about Vipul’s case is that he doesn’t use artificial intelligence as a crutch — he uses it as an amplifier for his own technical skills. He already has an engineering background, so he knows how to evaluate what AI generates and correct it when needed. This combination of technical knowledge plus powerful AI tools is what makes vibe coding truly effective. Without that context, the risk is accepting bad code without realizing it. With it, the result is a much more agile development cycle where productivity goes up in a visible and measurable way.
Arvind Jain: when AI becomes your coworker
Arvind Jain, co-founder and CEO of Glean, an AI-powered corporate search and workplace productivity platform, has an approach that says a lot about how our relationship with technology is changing. He told Business Insider that the tool he uses the most is actually Glean’s own product. But the most revealing part was how he described the mindset shift that happened.
According to Jain, his instinct has changed. Today, when facing any task he needs to get done, the first question he asks is: can Glean handle this? AI, in his words, has become his coworker. Not a tool he accesses every now and then, but a constant presence in his flow of decisions and daily tasks.
A practical example he shared illustrates this transformation perfectly. When he wanted to research a major cloud provider’s product launch to explore potential partnerships, the traditional path would be to reach out to the company’s CTO, who would then ask team members to write a proposal. That process would take about two weeks and eat up several people’s time.
Instead, Jain simply asked AI to generate the first draft of the proposal. The result came back in one minute, without taking up anyone’s time but his own. He made a point of saying he reads the generated material to check if it makes sense, but noted that most of the time, the quality is surprisingly good. We’re talking about going from weeks to minutes on a task that involves research, competitive analysis, and structuring partnership arguments.
The search for lost knowledge inside companies
Beyond generating proposals, Arvind also uses AI to organize and retrieve internal company information — which makes total sense given the product he’s building at Glean. He believes one of the biggest problems for growing companies is knowledge that gets lost in emails, documents, Slack conversations, and databases that nobody can find when they actually need it.
Artificial intelligence steps in as a layer of organization and context retrieval, helping entire teams work with more available information and less redundant effort. It’s a vision of productivity that goes beyond the individual and thinks about the collective. When an entire company can access accumulated knowledge quickly and in context, decisions become better informed and the time spent searching for information drops dramatically. 🔍
Daniel Yanisse: vibe coding for everyone, not just engineers
Daniel Yanisse, co-founder and CEO of Checkr, a company that uses AI to run background checks, brought a perspective that stands out for its inclusivity. He told Business Insider that he uses AI mainly from an engineering standpoint, and that his experience with vibe coding was transformative. In his words, it’s incredible — but you only understand the power of this approach when you actually try it.
What sets Yanisse’s vision apart is that he didn’t stop at his own use. He recognizes that people without a technical background can get stuck and feel discouraged when the code hits a wall and the app breaks. To address this, Checkr created the role of AI solutions engineers — professionals dedicated to helping non-technical employees who want to take advantage of vibe coding but get stuck at some point in the process.
This initiative shows a mature understanding that democratizing access to technology doesn’t just mean handing someone a tool and expecting them to figure it out. It requires support, guidance, and an environment where making mistakes during the learning process is acceptable and even encouraged.
A tool map for every experience level
Yanisse also shared his vibe coding tool recommendations, organized by level of technical experience. For people who have never written a single line of code, he suggests Lovable, a platform that makes building applications accessible even for complete beginners. Once users get a bit more comfortable with the technical environment, the recommendation is to move to Replit, which offers more flexibility and control.
For engineers, Yanisse’s favorite is Cursor, a code editor with integrated AI that lets you speed up development significantly. He also mentioned that he’s evaluating adding ClaudeCode as an additional layer for the company, expanding the arsenal of tools available to the engineering team.
This layered approach, with different tools for different profiles, is a sign that Checkr is thinking about AI adoption in a strategic and structured way — not just as an isolated experiment by the CEO. 🎯
What these three cases have in common
Looking at all three profiles, it’s clear that the way these startup CEOs use AI tools doesn’t follow a single pattern. Each one adapted the technology to their own context, to the specific pain points in their work, and to the decisions they need to make with more speed and confidence. Vipul focuses on prototyping, operational diagnostics, and contract analysis. Arvind turned AI into his main coworker for research and proposal generation. Daniel is democratizing vibe coding across the entire company.
There’s no universal recipe, but there is an important point of convergence: all of them treat AI as an active collaborator, not as a passive tool sitting around waiting for a specific prompt to function. The interaction is continuous, integrated into their workflow, and treated with the same naturalness as using email or a spreadsheet.
Human judgment remains at the center
Another point worth noting is that none of them abandoned human judgment. Vipul’s vibe coding still goes through his technical eye before it reaches the team. The proposals generated by Arvind’s AI are still read and reviewed before being sent out. The vibe coding efforts of Checkr’s non-technical employees have dedicated support engineers to ensure quality.
AI accelerates, organizes, and expands capabilities — but the final call remains human. This isn’t a limitation of the technology. It’s a conscious choice about how to integrate it into the workflow without giving up the quality and responsibility that leading a company demands.
What this means for the future of work with AI
What these three cases ultimately show is that artificial intelligence has already moved past being a conference topic or a distant-future talking point. It’s happening right now, in the daily routines of the people building some of the most relevant tech companies of the moment. And the most interesting part is that the highest-impact uses aren’t the most complex or sophisticated ones — they’re the most practical, the ones that solve real problems around scheduling, decision-making, and communication that any professional recognizes in their own day-to-day life.
For anyone following the world of technology and artificial intelligence, these accounts serve as a reliable barometer. If the very people building the technology are using it this way — integrating AI into everyday tasks like document analysis, code generation, and meeting preparation — it becomes increasingly clear that the practical adoption of these tools is no longer a question of whether it will happen, but of how each professional and each company will find the best way to bring them into their own context.
The main message from these three San Francisco CEOs is clear: start using it. Test different tools, find the points in your workflow that benefit the most from intelligent automation, and treat AI for what it is — an ally that amplifies what you already know how to do. The future of work with artificial intelligence isn’t on some distant horizon. It has already begun. 🚀
