5YF Episode #35: Ema Founder & CEO Surojit Chatterjee

Future Of Employees: Agent Workforce

subscribe and listen today

Today we dive into the future of AI employees.

The rise of AI agents in the enterprise isn’t just about streamlining workflows — it’s about rewiring how companies operate from the inside out. In this episode of the 5 Year Frontier, I sat down with Surojit Chatterjee, CEO of Ema, to explore a future where AI employees work alongside humans, collaborate with each other in real time, and increasingly take on the day-to-day tasks that once required entire departments.

Surojit brings one of the most accomplished product backgrounds in tech. He served as Chief Product Officer at Coinbase, led product at Flipkart to scale India’s e-commerce giant, and spent nearly a decade at Google leading product for mobile ads, commerce, and enterprise tools. Now, at Ema, he’s building a new category: universal AI employees — fully agentic systems that can be trained, configured, and deployed to operate across sales, HR, legal, support, and beyond. The goal? Not just to assist humans — but to meaningfully amplify what organizations are capable of.

We have built hundreds of agents. Those agents come together to build AI employees. And those AI employees talk to each other.

The conversation surfaced three transformative trends that will define how we work in collaboration with AI over the next five years:

My 5 Year Outlook:

• Agents introduce the 24/7/365 workday: Company output will skyrocket with always on agents.‍
• Tokenized micropayments power the agent economy: High-frequency agent interactions will require a new layer of low-cost, automated transactions.‍
• Developing economies will accelerate not be threatened by AI workforce: AI talent will fill critical skill gaps in healthcare, education, and services — unlocking leapfrog growth.

Curious? Read on as I unpack each below 👇🏼

Agents introduce the 24/7/365 workday

One AI employee, you get some productivity boost. You get another, a bit more. But when they start talking to each other — it’s one plus one equals eleven.

As every function in the enterprise begins to be supported by AI employees, we’re moving toward a world where departments continue working long after humans have clocked off. These agents aren’t bound by office hours, energy levels, or the limits of human attention. They reason, act, and collaborate in real time — creating a foundation for organizations that run continuously, 24 hours a day, 365 days a year.

In this model, the organization begins to resemble an iceberg: the visible tip is human output, but beneath the surface lies an enormous agentic workforce doing the heavy lifting. To quantify this shift: a typical employee works around 2,000 hours per year — but an AI agent running continuously can deliver 8,760 hours annually, a 4.4x uplift in potential output simply by being always on. Now multiply that across an entire company — and then across the global economy — and the productivity gains become staggering.

And when those agents begin collaborating across companies, executing tasks between systems in real time, we won’t just see more output — we’ll need to build new infrastructure for how they coordinate, transact, and exchange value.

Surojit Chatterjee of Ema

Ema, short for Enterprise Machine Assistant, is on a mission to reimagine how work gets done in large organizations by building “universal AI employees.” These aren’t just standalone chatbots — they’re sophisticated, mesh-like networks of specialized agents that can autonomously execute workflows across departments like HR, customer support, sales, and compliance. What sets Ema apart is its no-code, fully agentic platform — allowing non-technical users to configure, onboard, and manage AI employees using only natural language instructions. With over 150 pre-built agents and a proprietary ensemble model called EmaFusion that orchestrates over 100 large language models, the company is pushing the edge of what’s possible in enterprise AI. Ema last raised a $50M Series A led by Accel and has become a rapid riser in the AI landscape.

Surojit Chatterjee is Ema’s founder and CEO and has one of the best product resumes in tech. He was most recently Chief Product Officer at Coinbase, helping scale one of the most important companies in the crypto economy. Before that, he led product teams at Google for nearly a decade, overseeing products across Mobile Ads, Shopping, and Search, and earlier served as Chief Product Officer at Flipkart, where he helped build India’s leading e-commerce platform. In addition to building Ema, Surojit is also an active angel investor, backing startups like Udemy and Palantir. He holds a Master’s in Computer Science from SUNY Buffalo, and an MBA from MIT Sloan.

Tokenized micropayments power the agent economy

When agents start collaborating across company lines — negotiating, sourcing, executing tasks — they’ll need the autonomy to move money. While large transactions will continue to require human approval chains what is more interesting is the rise of potential micro-payments between agents.

Traditional invoicing and approvals won’t cut it in a machine-speed and high-volume agentic world. Instead, we’ll see the rise of tokenized micropayments: a new economic infrastructure purpose-built for AI. These could be the sourcing of a 3rd party data source or the run of an external agent to complete the objective of your own AI employee.

Agents may negotiate prices, manage procurement, and execute payments — all in real time. For that, we’ll need something like tokenized microtransactions.

This requires programmable, traceable, low-cost payment rails — and likely, a convergence between AI systems and crypto primitives like stablecoins or blockchain-based clearing.

One early signal: OpenAgents, an open-source project from AutoGPT contributors, has begun integrating crypto wallets into autonomous agent workflows. Imagine an agent coordinating a marketing campaign and paying a third-party AI designer for assets — all autonomously, in fractions of a cent. This isn’t theoretical — it’s already being prototyped.

As the economy of bots emerges, we must ask: who benefits the most from intelligent labor that’s affordable, scalable, and instantly deployable?

‍

Developing economies will accelerate not be threatened by AI workforce

If every child had their own personalized AI teacher, or every clinic had access to an AI medical assistant — the economic upside would be massive.

In many developing nations, the fear isn’t over-automation or losing offshore jobs to AI — it’s the persistent shortage of skilled labor needed to drive economic growth. From doctors and teachers to civil servants and administrators, the demand for quality services dramatically outstrips supply. AI employees offer a fast-track solution, giving these countries a chance to leapfrog traditional development pathways — not by replacing humans, but by filling critical gaps while the human workforce scales.

Surojit points to examples like a rural clinic gaining an AI medical assistant or a remote village student learning from a personalized AI tutor. These agents don’t replace people — they buy time, build capacity, and expand access. We’ve seen this story before: mobile phones leapfrogged landline infrastructure and unlocked new economic activity in the developing world. Now, the same pattern is playing out in labor. Startups across Latin America and Africa are already deploying AI-powered legal assistants, call centers, and virtual teachers. One Nigerian edtech company even launched a generative AI tutor capable of coaching students in native dialects — with no onboarding required.

And that brings us to the big picture. AI employees aren’t here to sit quietly in the corner — they’re here to work. As Surojit makes clear, they’ll operate across time zones, payment rails, and economic systems. The question is no longer whether enterprises will adopt agents — it’s whether they’re prepared for what happens when the agents start running the show.

Time to work.

‍