RamAIn builds the The world’s fastest computer usage agents for business work. We are a YC W26 company with a mission to eliminate manual and repetitive workflows by training AI agents that operate legacy systems, desktop applications and web portals the same way humans do, but 10 times faster and more reliably.
Founders
RamAIn was founded by Shourya Vir Jain (CEO) and Vansh Ramani (CTO), who met at IIT Delhi and dropped out to create native AI automation for enterprise workflows.
Shourya previously worked at McKinsey, where he saw firsthand how much business work still relies on manual interaction with legacy systems. He is also a FIDE ranked chess player (2118), ranked 20th in the world’s top 20 under 16s, and previously built and scaled an AI business company to over 6 figures in ARR.
Vansh is an AI researcher who worked at CMU on scalable machine learning and representation learning. His research includes publications in ICLR and ACS, and “Panaroma”, one of the fastest vector search algorithms, which was later merged with Meta’s FAISS. It focuses on creating high-performance planning and reasoning systems for real-world implementation.
Together, they founded RamAIn to create the fastest computing agents for enterprise work, combining deep research with production-grade systems that operate reliably in messy, real-world software environments.
Role
We are hiring a founding, early-career AI/ML researcher to join our core team and help build agents that autonomously reason, plan, and execute complex workflows. This role sits at the intersection of cutting-edge research and production systems: you won’t just prototype ideas, you’ll ship them to real enterprise customers.
You may be a good fit if you have a bachelor’s degree with exceptional practical experience building agent systems or have recently completed a PhD or master’s degree in machine learning, artificial intelligence, robotics or RL pipelines, multimodal modeling, or a related field.
What will you work on?
• Design of planning and reasoning architectures for long-term computer use agents.
• Training multi-modal models comprising UI layouts, screenshots and DOM structures.
• Build fast stock picking systems that operate through messy real-world software.
• Improve reliability, latency and robustness of deployed agents
• Run experiments and quickly ship improvements to production.
Expectations
This is not an exclusive research role. In the first few weeks, you’ll ship models, run experiments on real workflows, and see how your work automates production tasks used by business teams.
We are looking for someone who
• Move quickly from idea → experiment → implementation
• Created real AI agent orchestration systems.
• You are excited about agent AI, reasoning systems, experimentation or automation.
• Thrives in a rapidly evolving environment of early-stage startups
You will work directly with founders to push the frontier of IT agents and help define a new category of AI-native business automation.
