With AI, most of the work done in the digital world across functions can be automated today. Organizations that successfully automate routine processes get ten times more efficient and get more done with less. As McKinsey notes,“automation of activities can enable businesses to improve performance, by reducing errors and improving quality and speed, and in some cases achieving outcomes that go beyond human capabilities.” The productivity differential compounds over time. Those who fail to automate waste their time on tasks that add no differentiated value while others move faster with leaner operations.
This impact is clear especially in the case of engineers. With tools like Claude Code and Cursor, engineers are much more productive, doing ten times as much as before. This is also reflected in the substantial salaries and bonuses commanded by those who are proficient with the technology.
However for non engineering functions, the productivity revolution has yet to arrive. The business case is clear. What to automate is clear. However, today’s tools still require users to think programmatically. Visual drag and drop interfaces mask code, the underlying logic of triggers, conditional actions, and data transformations still require software experience. For non technical teams and individual professionals, the learning curve proves prohibitive. The result is a gap between potential and reality that has persisted for decades. A recent MIT study found that “95% of corporate generative AI pilot projects are failing to deliver measurable financial returns” highlighting just how wide this adoption gap remains.
Work Atoms, founded by Mr. Dhruv Jaglan and Mr. Georgi Boby, is building Rilo, a platform that approaches automation differently. Users create workflows by describing tasks in natural language or demonstrating them directly. They’re building a world where knowing what to automate is enough to get it done. No code. No technical bottlenecks. They’re initially focusing on Sales and Marketing function automations as these comprise more than 50% of use cases.
Enterprise Scale Experience
Mr. Jaglan’s background spans enterprise software and artificial intelligence at significant scale. After graduating from the Indian Institute of Technology Bombay with a degree in computer science, he joined LinkedIn as a software engineer, working on infrastructure that served hundreds of millions of users globally. The experience established his understanding of what reliability requires when systems operate at scale.
He subsequently served as Cofounder and Chief Technology Officer at Babblebots, where he led development of AI powered interview systems. The platform conducted voice based candidate assessments in more than twenty languages, processing hundreds of interviews simultaneously using proprietary deep learning models. Notably, this work was completed before the launch of ChatGPT, at a time when popular AI systems struggled to perform even basic language tasks reliably. Enterprise clients used the system for high volume hiring, compressing timelines from weeks to hours while maintaining assessment quality across diverse linguistic and cultural contexts. The systems had to perform reliably at scale because hiring decisions carry significant consequences for both employers and candidates.
The work earned Mr. Jaglan recognition from Sigma Square as one of their 25 Under 25 Entrepreneurs Worldwide. Earlier, his research at the Technical University of Braunschweig on 3D reconstruction algorithms for medical applications resulted in peer reviewed publication, demonstrating capability for rigorous technical contribution across domains.
“The HR teams using our platform knew exactly what they needed,” Mr. Jaglan said. “They understood the assessment criteria, the interview flow, the evaluation framework. But they could not have built the system themselves. That gap between domain expertise and technical capability exists across most enterprise functions.”
Robotics and Observational Learning
Mr. Boby’s background is in robotics and machine learning. He completed his electrical engineering degree at IIT Bombay in three years, finishing a year ahead of schedule. Upon graduation, he joined CloudChef as its first engineer, tasked with building robotic systems that could replicate cooking techniques from professional chefs in Michelin starred establishments.
The technical challenge required rethinking how robots acquire capabilities. Rather than programming specific recipes and movements, Mr. Boby developed systems that interpreted chef movements and processes using machine learning models that translated observation into robotic action. The systems learned by watching. CloudChef’s technology, backed by more than $20 million from top investors, now operates in multiple Michelin-starred restaurants and airline kitchens across the United States.
“At Cloud Chef, we built systems that learned by watching,” Mr. Boby said. “The chef demonstrates, the robot observes and replicates. If you can automate cooking, which is one of the most unstructured environments, automating sales and admin is a piece of cake. The same observational learning principle applies to workflow automation.”
Addressing the Adoption Gap
Work Atoms represents an attempt to address the gap between automation potential and automation reality.The people who understand a workflow rarely know how to automate it. And the people who can automate it rarely understand the workflow. The challenge is significant: “only 4% of businesses have fully automated their workflows, showing a large gap between potential and current adoption.“ Work Atoms represents an attempt to address the gap between automation potential and automation reality. Despite the clear business case, only 4% of businesses have fully automated their workflows, a gap that persists largely because the people who understand a workflow rarely have the technical skills to automate it. Rilo is a platform helping knowledge workers who understand their processes automate their work, without needing to write code, hire engineers, or think programmatically. Mr. Jaglan’s experience at Babblebots demonstrated that sophisticated AI could be made accessible to non-technical users when designed around their workflows. Mr. Boby’s work at CloudChef proved that systems could learn complex tasks through observation rather than explicit programming. Rilo brings both principles together, making automation as simple as describing what needs to be done.
The company raised 1.5 million dollars in seed funding led by Peak XV Partners, with participation from De VC, Day Zero Ventures, and Mr. Gautam Prakash. Initial focus is on sales and marketing workflows, where repetitive tasks are well defined and the volume justifies automation investment. The strategy allows the company to prove its technology in environments where the consequences of errors are manageable before expanding to higher stakes functions.
The automation market remains competitive, with established players and well funded startups pursuing similar objectives. Whether Work Atoms can differentiate on reliability and accessibility will determine whether they can capture a meaningful share of an industry where many have promised transformation and few have delivered. The opportunity is substantial for whoever solves the problem. The challenge is equally significant.