ProfileCity

After over a year of building ProfileCity, Aman and I have decided to call it a wrap. We set out to build the first open-source ATS software. The idea was to make hiring, managing, and shortlisting candidates seamless—powered by AI agents, at zero setup costs. We started as a search engine to find people on the internet. But we quickly learned that recruiters were our real customers. So we pivoted. The search engine became the core of our ATS, designed to integrate into hiring workflows. We built AI agents that could scan thousands of profiles to surface the best candidates—fast and accurately. Early pilots with recruiters gave us great feedback, and for a while, the momentum was exciting. So, why are we shutting it down? Hard truths: 1. Open source didn’t help as much as we expected. At such an early stage, the community wasn’t big enough to drive adoption or contributions—it felt more like a distraction than an advantage. 2. Revenue models are tricky in HR tech. We underestimated how long it would take to get paying customers and how heavily some businesses rely on entrenched solutions. 3. Burn Out. Building in a feature-heavy, fast-moving space took a toll on us. Juggling the complexity without clear traction drained our energy and focus. But we learned a lot - 1. Speed is king. Mistakes are inevitable in startups, but moving fast means you can learn, fix, and build momentum. Slow progress kills motivation and opportunities. 2. Educate your Customer. AI isn’t just a tool—it’s a collaborator. Users need to understand how to trust and integrate AI into their workflows. And people adopt what they understand, not what's powerful. Thus, a great UX and ongoing user education are essential. 3. Focus is everything. Fancy features don’t make an MVP. Build what matters, ignore the noise. Every feature you skip is time you save for what truly matters. Talking directly to customers is non-negotiable. 4. Be brutally honest with yourself. Not every great idea will work. Identify the gaps early—whether it’s market fit, timing, or passion—and make decisions fast. You’ll save time, energy, and money. We’re transitioning ProfileCity to an MIT-licensed project. Maybe someone out there will take it further. What’s next for me? I’ve built startups before. It’s been a mix of challenges, fun, and learning. I’m exploring what’s next and am open to opportunities. If you’re building something exciting—I’d love to connect. My current focus is on software and GenAI applications. I bring a strong sense of agency and thrive in roles where I can access ownership. DMs are open, feel free to reach out. Lastly, huge thanks to: 1. Aman Desai, my co-founder—it’s been a privilege building alongside. 2. NVIDIA, OpenAI, and AWS for the grants and infrastructure that powered us. 3. Professor Nik Bear Brown for your guidance and support. 4. Devansh Shah, Dhruv Daftary, and Shubham Jain for being there throughout.

We recently released Vidur V1, and we just convinced a large recruiting firm from Boston to do a pilot with us. I believe we’re creating something fundamentally groundbreaking in recruiting. Here’s why: 1. Vidur is open-source - You can clone it, use it for free, and only upgrade when you genuinely need more. It flips the traditional licensing model on its head, especially in legacy ORMs (viz., Workday) 2. Our approach to candidate shortlisting is deeply contextual. Think of it as Perplexity versus Internet Explorer — smart, adaptive, and modern. The system understands the nuances of hiring decisions like never before. 3. Our agentic UX is a hiring co-pilot. Let Vidur know who you are hiring, and it constructs a customizable hiring pipeline — interviews, assessments, and third-party integrations included. It’s like the Future of Work, NOW! We’re in Beta, moving at lightning speed. We can help you if: 1. You’re hiring. 2. You run a recruiting agency. 3. You have a resume dump and need intelligence to find talent. We’re looking for contributors and developers with a passion for reshaping the future of recruiting. There’s something powerful about building the systems that will define how companies will hire with intention. Links in the comments. Follow ProfileCity.

What began as a simple project has turned into the top open-source recruiting software on GitHub! If you search for "Vidur," "Recruiting," or "Recruiting Software" on GitHub, we’re ranked number one, even ahead of Microsoft's repo! A huge thank you to everyone who has starred and contributed to our repo. If you're looking to grow as a software developer, contributing to Vidur is an amazing opportunity. Drop your email in the comments, and we’ll send you the onboarding details! More details in the first comment! Follow ProfileCity!

Hey everyone! I’m super excited to announce that me, Aman Desai, and the whole team at ProfileCity are open-sourcing the first-ever recruiting software on GitHub. Today, we’re officially releasing Vidur's first stable version, built using Nuxt. Vidur gives companies full control over their recruiting process. We’ve already started work on V2.0.0 with even bigger plans: - Customizable hiring workflows - AI agents to assist in recruiting - Plugins to connect with third-party tools - Full support for deploying backends on Supabase and Appwrite - Deploy on Vercel, Docker, AWS, and Azure We’re also launching a closed-source enterprise service with: - AI-powered candidate profiles - Recruiting assistants to help you hire smarter We’re building the best UX on recruiting software ever. If you’re a startup using Google Forms or Notion to hire, let’s chat! We've got exciting things planned! We're also hiring software and cloud engineers to push our open-source initiatives. A special thanks to our community collaborators - Shivam Sharma, Akash Varun, Hemal Desai, PRAMATH S, Harshil Shah, Anirban Khara, Sohni Rais, Hardik Jain, Dhanush G, Pushkar Mishra, Devansh Shah!

🚀 Calling for Software and DevOps Interns! 🚀 We’re recruiting Software and DevOps interns to push our open-source efforts to the next level with the official Vidur v1 release! Vidur is the first-ever open-source recruiting software, and we’re growing faster than ever! Apply here: https://lnkd.in/eXFFTJRT Here’s why you should join us: - We’ve already hit 180+ stars on GitHub (and rocketing!). - We break things, move fast, and build even faster—the community loves it! - You’ll connect with awesome developers, founders, and recruiters while building your network and sharpening your skills. It’s All About: - Remote & Flexible: Work from anywhere, at your own pace. - Fun & Chill Vibes: No corporate nonsense, just a cool project with awesome people. - Open-source Culture: Your contributions get recognized, and you can shape the future of recruiting software. Join Dev Shah and Aman Desai in building Vidur! (other details in the first comment)

🚀 Calling for Software and DevOps Interns! 🚀 We’re pushing our open-source efforts to the next level with the official Vidur v1 release! Vidur is the first-ever open-source recruiting software, and we’re growing faster than ever! Apply here: https://lnkd.in/eGCFrh7j Here’s why you should join us: - We’ve already hit 180+ stars on GitHub (and rocketing!). - We break things, move fast, and build even faster—the community loves it! - You’ll connect with awesome developers, founders, and recruiters while building your network and sharpening your skills. It’s All About: - Remote & Flexible: Work from anywhere, at your own pace. - Fun & Chill Vibes: No corporate nonsense, just a cool project with awesome people. - Open-source Culture: Your contributions get recognized, and you can shape the future of recruiting software. Join Me and Aman Desai in building Vidur! (other details in the first comment)

open source, ftw! 🚀 comment your email address and we'll send you an onboarding to contribute to our V2!

our open-source recruiting software just hit 150+ stars and 10+ contributors on github! yayy! 🥳 if you are an early-stage company, using google forms or notion to hire, hmu! we're cooking! if you wanna join our awesome crew and contribute to open-source, comment "interested" and will slide you our discord invite! back to building cool things! 🐋 . . . follow ProfileCity (previously Nirvana Labs) would really appreciate if you'd ⭐ the repo! link in the first comment!

6k applications for a frontend engineering role in 48 hours. How do you plan to standout? The hard truth is: If you're applying to 100s of jobs and hearing nothing back, it might be time for some self-reflection. My founder friends are hiring and what I've learnt from them is eye-opening. In less than a year, how resumes are screened, read, and ranked has evolved into the "Proof of Work" filter. With the GPT effect, building products has never been easier, and that is both exciting and terrifying. It means the bar has been raised. A degree in Engineering, a leetcode score, or a course certificate are all table stakes now. Everyone's got 'em. This is what I did and what you should do too - 1. Polish your persona. The way you express yourself, and the ideas you share - make them uniquely you. It is easy to copy-paste with GPT, which makes Authencity the King. 2. Get your hands dirty. Contribute to real projects. Master your craft until it becomes an extension of yourself. That's when the magic happens. 3. Build connections like your career depends on it - because it does. Nurture relationships. Be genuine. Remember, your network is your net worth. Though the game has changed, the core rule remains: Less is more. Simple. ps - 16% seniors level people 💀

Hey LinkedIn Fam, after almost a year, we’ve finally decided to shut down Nirvana Labs. When we started Nirvana Labs last October, we were just a bunch of passionate folks building the most powerful People’s Search. It was ambitious, and it was daring. By January, we had our first taste of success. Our prototype was live, with 8,000 users already indexed. The excitement was palpable. We could feel the potential humming through the veins. But the startup world is nothing if not unpredictable. Our soft launch brought a harsh reality: users liked our product but as a toy. They didn't really need it. It was a gut punch but it taught us some invaluable lessons: 1. Building an aspirational professional product is tougher than making a professional product aspirational. 2. Users who like your product but don't need it won't become repeat customers. That's crucial for product-market fit, also sustainability remains uncertain. For the next few months, we didn’t ship any major product updates but instead iterated over various new user bases. Ultimately, targeting recruiters and early-stage companies. This is when we struck gold. They loved it. They needed it. But they also wanted more – they wanted our search engine to work with their candidate data. They needed our AI agents to work with their ATS systems. At that moment, we faced a choice. We could stubbornly cling to our original vision, or we could listen to our customers and build what they truly wanted. We chose the latter. The market spoke and we heard. “Don’t give up. Pivot.” And so, Nirvana Labs is now rebranding to ProfileCity (previously Nirvana Labs). We're shutting down all our consumer-facing operations to double down on what our customers need. We’re building Vidur, the first-ever open-source Recruiting OS. The search engine that was once the heart of Nirvana Labs now beats at the core of Vidur's ATS system. To all the startups, early-stage companies, and recruiters out there: we've got something special cooking for you. DMs open! Check out our repo, and drop us a star - every star helps us grow faster. (link in the first comment) Welcome to Profilecity!