Sprinklr · Agent Assist
Quietly Powerful — An Invisible Ally
Complex support cases overwhelmed agents mid-conversation. I designed the AI that quietly does the busywork — without taking the wheel.
The context
What you're looking at
Sprinklr
Sprinklr is an enterprise customer-experience platform — large brands use it to run customer service, social, and marketing across every channel from one place. This work lives in Sprinklr Service, the contact-center product where support agents handle live customer conversations all day.
Agent Assist — where I worked
Agent Assist is Sprinklr’s AI layer for live customer service, spanning smart responses, compliance checks, and sentiment nudges. Within it, I owned Guided Workflows — and extended the pattern into the admin builder teams use to create those flows without engineering.
Overview
An AI that does the busywork — not the deciding
Agent Assist is an AI workflow system built into Sprinklr’s Case suite. It handles the repetitive backend work — pulling data, updating tickets, surfacing the next best action — in real time, right inside the live conversation. The principle underneath the design: automate the busywork, but leave every decision with the agent.
Impact
What it moved
What’s mine:I designed the patterns; Product and Data tracked these across rollout. Efficiency is aligned to Sprinklr’s published benchmarks, CSAT came from early-rollout feedback, and the error drop from QA logs. The design was mine; the measurement was a team effort.
The challenge
Live support kept breaking mid-conversation
Agents couldn’t see their own progress, juggled multiple active workflows, and leaned on unclear instructions — so complex cases meant cognitive load, delays, and errors. And admins couldn’t change a workflow without engineering, so the system couldn’t keep pace with the business.
I'd lose my place mid-chat while switching tabs — it broke my rhythm with the customer.
Even small workflow changes needed engineering. It slowed us down every time.
The question
How might we streamline complex workflows so support teams focus on customers — not tools?
Insight
Where AI should step in — and where it shouldn't
Not every problem needed a feature; some needed quiet intelligence. AI belonged exactly where repetition and hesitation lived — auto-filling data, updating tickets, surfacing the next-best action. But the moment it started deciding for the agent, it would cost the one thing live support runs on: trust. That boundary became the product.
Key design decisions
The calls that shaped it
The hardest back-and-forth was on the guided workflow itself. The push was for more AI, and something flashier than the competition — but the timeline ruled that out, and more AI meant less room for the agent to step in. An early version laid the steps out horizontally; it crowded the panel and didn’t scale, so I moved to a guided stepper— one step at a time, where the agent’s choice at each step drives what surfaces next. The AI handles what comes next; the agent stays in control of every call. A familiar, predictable shape can feel like settling next to something flashier. It wasn’t — under live pressure, predictable-and-in-control is exactly what an agent wants.
The solution
One system, two audiences
Agent-facing
AI-powered Guided Workflows
Real-time guidance rendered as structured, step-by-step flows the agent completes inside the conversation — no tab-switching, no lost place.
Admin-facing
No-code Workflow Builder
Admins design, style, and deploy AI-assisted flows themselves — removing the engineering dependency and letting the system iterate at the speed of the business.
Agent-facing · the Care Console
- 1Guidance lives inside the conversation — the agent never leaves the chat.
- 2Recommended Actions are suggested; the agent confirms — AI assists, never executes.
- 3A progress stepper for wayfinding — what's done, the step they're on, and what's left, all at a glance — so the agent never loses their place in a multi-step task.
Admin-facing · the Guided Path Builder
- 1A drag-and-drop component library — checkboxes, radios, inputs, ready to place.
- 2Publish with no engineering — admins ship changes themselves.
- 3Live preview shows exactly what the agent will see.
- 4Drag any component straight onto the canvas.
Explorations
A few directions before it clicked
Working straight in high-fidelity, I pressure-tested real alternatives against feedback from daily ticket reviews before the final pattern landed.
Sentiment, trust & compliance
From what agents did to how they said it
As the AI matured, the focus expanded from what agents did to how they said it. I partnered with product and engineering on real-time sentiment and compliance guidance — defining when and how feedback surfaced as agents typed, without breaking their flow. This work shipped into Sprinklr’s AI Service platform.
One line, rewritten
“Warning: message flagged for tone.”
Reads like being policed.“This might land as a little abrupt — soften it before you send?”
Reads like a teammate looking out for you.
The words are the product — clear microcopy did more work than heavier UI ever could.
Reflection
Calm, on purpose
This project was about making a dense product feel lighter for the people who live in it all day. Three things stuck:
- Designing for AI is designing for trust— the interface’s real job was to make an invisible, probabilistic system feel predictable and accountable, so agents would actually lean on it instead of second-guessing it.
- Accessibility reshaped it — small contrast and sizing changes made workflows calmer and faster.
- Words outperformed features — clear microcopy unlocked more ease than heavy UI ever did.
- Systems thinking beat screen thinking — designing for agents and admins forced me to build for scale, not just a screen.
Next: deeper usability testing, a UX-writing style guide, and bringing richer accessibility — screen-reader support — into the product.