Nitya Kumar, an Adobe employee in India, says she swapped the usual “climb the ladder” mindset for a science-lab approach to learning AI—experimenting daily, building prototypes, and turning quirky ideas into interview stories that helped her land an AI-relate
In 2022, Nitya Kumar walked into a product design job at Meta straight out of college. In 2024, she watched the industry pivot toward AI—and decided she wasn’t going to wait until she “felt ready” to learn it.
Kumar, 25 and based in India, describes her turning point as a shift in how she saw her career. She stopped treating it like a ladder where the goal was simply moving upward. “All work and no play keeps the Big Tech offers away. ” she said. and she acted on that belief by treating her career like a science lab—giving herself room to experiment rather than chase perfection.
The move mattered because she didn’t come to AI with a traditional machine learning background. She went to art school in the US and says she built her AI education herself. using YouTube videos and learning from friends. Rather than trying to master everything at once. she focused on experimentation—especially with Cursor—and she recruited friends to help keep her accountable.
She describes the method like a formula: one hour of Cursor a day plus seven days to produce a “functional AI prototype.” By the end of that first week. she had created a game that could detect and track the gestures of a user dancing in their room. What started as a fear of learning a new technology turned into confidence through doing.
Kumar’s experiments grew into a workflow she calls “vibe coding,” using different tools for different parts of a project. She began with Gemini and Cursor, then gradually expanded to other tools, including Figma MCP. She also stopped sticking only to safe case studies. One experiment generated Matcha recipes. For another build. she used Gemini to develop and refine prompts that she would feed into Claude to vibe-code the product.
Those projects weren’t just practice. Kumar says they helped her stand out in interviews, in a very human way: they made interviewers laugh, and conversations shifted toward her experiments—debating matcha flavors while she played around with the tools she’d built.
After roughly four months of intentional AI experimentation, Kumar landed a UX design job at Adobe. She started in November 2025, and she says the role involves developing agentic AI experiences, including conversational AI assistants.
Once she arrived, she didn’t stop learning. She says she still had skills she wanted to build, but her earlier experimentation gave her confidence to do more than just work. After around three months at Adobe, she decided she wanted to share what she’d learned with the design team.
Every other Friday, she leads an AI playground workshop for other designers at Adobe in India. The sessions run online. The group experiments with new tools and shares how AI-assisted workflows fit into day-to-day work. They also collaborate on staying ahead of a “technical curve” that Kumar says designers across the team are facing.
Designers come to her for help with debugging problems and building prototypes with AI. As she supports them, her toolkit and confidence grow as well. In her view. treating her career like a science lab didn’t just help her transition into an AI role without a machine learning background—it helped her succeed inside it. and find a way to master vibe coding while building creativity and leadership skills at the same time.
Kumar’s story lands on a simple insistence: fear can be beaten, but not by waiting. For her, it was beaten by making learning a daily experiment—and then bringing the results into the work and the room where colleagues decide what to build next.
AI design vibe coding Cursor Gemini Claude Figma MCP Adobe Meta UX design agentic AI conversational assistants India tech jobs
Cursor every day… sounds like a job filter lol.
So she basically gamed it by doing AI like a lab? I feel like everyone says you just need YouTube and friends but that’s not realistic for regular people. Also she already worked at Meta right? lucky timing.
Wait she says “all work and no play keeps the Big Tech offers away” but the whole thing is still grinding 1 hour a day and prototypes. Like isn’t that work? Idk I watched one video and now I’m an expert now apparently. Also gesture-detecting game sounds fake like, no way it worked that fast.
This is kinda the first time I’ve heard someone admit they don’t have a real ML background and still got in. But also art school in the US + friends + Cursor + Gemini… that’s like a whole ecosystem, not just “learn yourself.” I’m sure the interview story helped, but Big Tech is always changing what they want anyway. Not saying it won’t work, just feels like everyone leaves out the part where they already had access to stuff.