Hi, I'm Nishtha.

I lead design and research at Workato, where I'm building the platform that gives a flock of unpredictable AI agents their rules and shape.

Platform strategy

Shaping an integrated AI platform strategy

Used research and prototypes to win support for an integrated platform over a standalone product, then guided its launch and what came next.

Product incubation

Translating the MCP standard into a product

I understood the emerging MCP standard, translated it into a product model, and guided its implementation from prototype through release.

Design org leadership

Scaling a design org from seed-stage to a $5.7B company

Built Design and Research from one to 22 over ten years. Created shared practices and design leaders to keep the quality bar high.

Design direction

Supervising AI agents without leaving your work

Led a product-wide model for monitoring and steering parallel AI agents without interrupting people’s current work.

Hands-on design

Designing workflow primitives that scaled into a platform

Designed composable workflow primitives that let customers build almost anything and later supported copilots, agent skills, and MCP.

Co-inventor

Patenting a turn-taking cue for streaming AI

I co-invented a patented conversational cue that shows when streaming AI is still responding and when the user can take the next turn.

Some tinkering. A few musings.

Trust in human-robot interaction

5 levels of AI maturity in design

Design your own bird swarm

Nishtha presenting on stage at a Workato conference

I've been leading Product Design and Research at Workato for the last 10 years.

With two master's and a bachelor's degree, my formal training spans physical product design, visual communication, psychology, and human factors.

Before Workato, I co-founded MyNewsStudio. We built a good product, but the business model was not viable. We also tried to do too much at once. The experience showed me that design alone cannot create success. Product strategy, revenue model, sales, distribution, and marketing all have to work together. Focus is a business decision. We eventually wound down the company, so knowing when to walk away was my hardest lesson.

Before software, I designed tabletop accessories for Episode, an Indian design brand. Sixteen years later, elements of that work are still sold and used to market the collection. I've also created art installations, designed luxury retail spaces, and worked as a design merchandiser.

I'm looking to build with companies, in software or hardware, where the hard part is making powerful systems people can trust.

LinkedIn comment from Vijay Tella, Co-founder and CEO of Workato: “Nishtha Mehrotra, I'm incredibly grateful. We're so lucky to have you leading the intuitive product experience we're known for. I've spent 30 years in integration middleware, and this tech is complex and whack under the hood. Refactoring six complex platform categories into one while making it seamless for both pro-code integration experts and business teams? No joke. What you and your team have built is one of the great feats of enterprise software design. So much so, you have created a marketing challenge as Workato confounds people. How can something that looks so easy be so powerful? What blows me away is your first-principles thinking and relentless drive to keep attacking the complexity. We don't have real competition for our product experience, productivity, and speed; our biggest challenger is ourselves. You and your team are awesome role models of our core value to keep raising our game. Thank you.”

From Vijay Tella, Workato’s CEO

Credentials

M.S. in Human Factors and Ergonomics

Department of Industrial and Systems Engineering San José State University

M.Design

Indian Institute of Technology Guwahati, India

B.Design

National Institute of Fashion Technology, Gandhinagar, India

Supervising AI agents without leaving your work

TL;DR

  1. Context

    In 2026, the market is shifting from copilots to autonomous agents. Workato already lets people spin up multiple agents to work in parallel across the product. But you cannot tell from the interface because it still looks like a single AI assistant.

  2. Objective

    Make parallel agent work visible while letting people stay focused on their own tasks.

  3. Action

    Treated agent activity as a layer that followed the user, rather than something tied to a page or workspace. Built the first prototype and directed its evolution into a persistent bottom bar.

  4. Outcome

    Created one place to see what every agent was doing, step in when needed, return to any session, or queue more work. The design is now under development.

How do you supervise agents without becoming their full-time supervisor?

Workato’s AI appeared as a single copilot inside a chat panel. But the product could support several agents working in parallel. We needed to make that capability more visible. That created a new problem: how could people supervise all those agents without repeatedly stopping their work or switching context?

People need to assign work, leave agents running, and continue with their own tasks. At the same time, they need to know when an agent finishes, gets stuck, or requires a decision.

Imagine an automation lead starting three agents:

  • One debugging a failed invoice sync in Finance
  • One building a lead-routing workflow in Sales
  • One investigating errors in an employee-onboarding process

Rather than waiting, the automation lead begins building an approval workflow manually. The agents continue working in the background while the lead remains free to check their progress, redirect them, or queue additional work. The automation lead only needs to return to the full agent session when more context is required.

Why is this particularly difficult in Workato?

An enterprise automation lead may support several business teams, such as Revenue Operations, Procurement, and People Operations. Each team can have its own workspace, data region, permissions, and development, test, and production environments. Some environments have several active agents, some have none, and others contain data the lead can't access.

An agent could therefore be working in a different workspace, environment, or product area from the page the user is currently viewing. If agent supervision lived inside a page or asset-level chat, people would have to remember where each agent started and navigate back to that location to check its progress.

The design challenge was to make agent supervision independent of any asset, page, workspace, or environment. It needed to belong to the user, preserve each agent's context, respect permissions, and remain accessible throughout the product.

Example of Workato's complexity showing one enterprise automation lead supervising agents across Revenue Operations, Procurement and Supply Chain, and People Operations workspaces, each with development, test, and production environments, different agent states, and permission boundaries

One automation lead may supervise agents across business workspaces and environments, while permissions determine which activity is visible.

The first concept confused location with attention

An early concept combined agent status with the workspace and environment indicators.

I pushed back on this direction during critique. The environment switcher answers, “Where am I working?” The agent indicator needs to answer, “Does an agent need my attention?” Combining the two suggested that agent activity belonged only to the current environment.

Early agent control combined with the environment switcher, alongside Figma feedback explaining that agent supervision needed to work across workspaces and environments
My feedback in Figma
Can AIRO occupy its own layer across the product? If I have agents working in workspace A and workspace B, I shouldn't have to switch workspaces just to see what is happening.
The environment switcher already communicated location. Agent supervision needed a separate, product-wide layer.

Making the principle tangible

I used our internal product scaffolding and Claude Code to build a very quick and dirty prototype. It gave Serene and me something concrete to evaluate and made three interaction principles visible:

  • Running agents should remain accessible throughout the product.
  • Their status shouldn't depend on the current workspace or environment.
  • Anything requiring a decision should be reachable in one action.

Finding the right edge of the screen

We then explored four placements.

A full-width top bar

Our first direction placed a bar across the top of the product. It was directly inspired by the prototype I had built. This created more room for active agents, but it made the top of the screen feel heavy and pushed the page content down.

The hierarchy also became unclear. Because the bar spanned every workspace and environment and sat above profile settings, it appeared to sit above the entire product. Agent supervision and profile settings are parallel product-wide layers. An upcoming control plane would need the same architectural territory.

The bar also pushed the workspace and environment switchers away from the screen edge and into the middle. Edge targets are easier to find and reach, and we didn't want agent supervision to weaken this important navigation.

Full-width agent bar positioned above the workspace and environment switchers at the top of the Workato interface

A centered notch

We then condensed the bar into a centered notch inspired by the MacBook notch. It could expand to provide a quick view of active agents without introducing a permanent panel.

However, it collided with an existing control in the recipe editor, where people switch between Edit and Test modes.

Centered agent island at the top of the Workato interface showing agents that need attention, agents that are ready, and an action to start another agent
Expanded centered agent island colliding with the Build and Test controls in the full-screen recipe editor

The right edge of the screen

Next, we explored a persistent rail on the right, where the chat interface already opened. That proximity created a strong relationship between the agent overview and each conversation.

However, horizontal space was already limited, especially on smaller screens. A persistent rail alongside an open chat would leave too little room for the page itself. The narrow rail also forced agent groups and states into short, single-word labels. We wanted to use descriptive sentences that felt conversational, and the sidebar didn't give them enough room.

Agent overview with several parallel tasks and an expanded agent conversation in a persistent panel on the right side of the product

The bottom of the viewport

Finally, Serene moved the interaction to the bottom of the screen. This resolved several problems at once:

  • It avoided the existing navigation and mode controls.
  • It could appear on any page without restructuring that page.
  • It remained in peripheral view without competing with the current task.
  • It created a natural transition into a detailed agent view.
  • The action for starting an agent could sit close to the conversation that followed.
Bottom-position exploration showing the compact agent bar, parallel agent status, and an expanded supervision view without displacing the current Workato page
The bottom placement worked consistently across the product while allowing the interface to grow when the user needed more context.

Every agent, always within reach

The final agent bar stays at the bottom of the viewport throughout Workato. It doesn't belong to the current page, workspace, or environment. It belongs to the user.

From anywhere in the product, people can:

  • See which agents are active
  • Check progress without changing context
  • Respond when an agent needs input
  • Redirect an agent
  • Return to any agent session
  • Queue or start additional work
The bar remains unobtrusive while agents are working, but keeps their activity visible at a glance. When the user asks for more information, it expands into a richer view or opens the complete agent experience. The user controls when agent supervision occupies more of the screen.
I think of this as a “peek-a-boo” pattern: the interface stays at the edge of attention until there is something worth seeing, then recedes when the user returns to their work.
The interaction can move from a quiet status signal to full supervision, depending on how much context the user needs.

My contribution

I reframed the problem as a product-wide information architecture challenge and redirected an early concept that tied agent activity to the current environment. I established the core interaction principles and built the first prototype to communicate the direction. I also critiqued subsequent concepts and helped guide the design as it developed.

Serene led the detailed interaction design and identified the bottom placement that made the pattern work across the product.