Description
Agora is a global leader in real-time engagement providing developers with simple, flexible, and powerful APIs to embed real-time conversational AI, voice, video, interactive live streaming, and chat into their applications and IoT devices.
In October 2024, Agora launched its Conversational AI SDK, integrated with OpenAI's Realtime API as a first step the direction towards AI. Partnering with OpenAI was a game-changer, bringing in major traction.We knew we had to move fast to capitalize on the momentum. Since then, we’ve designed and shipped multiple foundational products within the Conversational AI Suite.
These initiatives have directly accelerated revenue growth and customer adoption, amplified industry visibility, and built sustained market momentum.
TEAM
Sid Sharma, Vice President
Kevin Tu, Product Manager
Jolene Zhang, Product Manager
Shakti Soni , Principal Product Designer
8+ Developers
Timeline
2024-25
Key Projects
This caselet highlights some of the key products I worked on as a part of the Conversational AI Suite. If you're interested in hearing about any of them in detail, reach out to me.
Description
A cross-platform demo app (Web, iOS, Android) showcasing real-time, low-latency conversational AI agents built via Agora’s Conversational AI Engine.
Live link
Overview
After launching its Conversational AI SDK in collaboration OpenAI in October 2024, Agora saw a surge of interest and demo requests. Explaining the SDK through documentation alone wasn’t enough to convert interest into adoption.
Goal
Businesses found it difficult to visualise the SDK’s real-world applications across platforms without a tangible example. The goal was to demonstrate the SDK’s utility quickly and effectively.
My Role
The demo app was an all hands on deck project. I worked side-by-side with the lead designer to design the entire experience end-to-end, working in rapid design sprints to deliver quickly under tight timelines.
Solution
A cross-platform demo app (Web, iOS, Android) showcasing real-time, low-latency conversational AI agents built via Agora’s Conversational AI Engine.
Accelerated Adoption: 300% increase in inbound requests after demo launch.
Faster Sales Cycle: Prospects could experience the SDK firsthand, reducing technical onboarding time by 40%.
Brand Credibility: Strengthened Agora’s image as a leader innovating in real-time conversational AI.
Description
A platform that tests and compares different AI voice models and stack of AI models (ASR-LLM-TTS) in real-world conditions, helping developers find which ones work the fastest and sound the best in conversations.
Live link
Goal
Enable businesses to easily evaluate and compare the performance of leading AI models in real-world conditions before integrating them into voice applications.
My Role
Led design for 3 of 5 core pages and supported the rest to maintain consistency. Collaborated with PMs for alignment, led design QA, and built a tokenized Figma UI kit with 30+ custom components and full dark/light mode support.
Solution
The lab doesn’t just highlight Agora’s engine — it orchestrates the ecosystem of AI models around it. This positions Agora as the neutral, enabling layer for scalable Conversational AI, rather than just another tool vendor.By launching an industry-first initiative like this, Agora showcases itself as a leader shaping standards in Conversational AI performance, not just a follower.
Ecosystem Growth → Benchmarked over 25 leading AI models across 10 regions, positioning Agora as the go-to hub for model evaluation and deployment.
Revenue Impact → Contributed to a 20% increase in Conversational AI-related revenue within the first year by driving adoption and upsell opportunities.
Description
A voice agent management dashboard and an integrated graph editor (TEN Super Playground) for creating, editing, and deploying agent flows.
Live link
Coming Soon!
Goal
Developers and businesses increasingly need a unified, flexible platform to design, test, and deploy conversational AI agents without juggling multiple tools, SDKs, and services.
The goal was to provide a unified “Super Playground” by merging Conversational AI Engine* with TEN*, combining the flexibility to experiment with the simplicity of rapid deployment.
Context
TEN is an open-source framework for building custom, real-time, multimodal conversational AI agents from scratch.
Conversational AI Engine is a hosted, low latency voice AI engine that lets you instantly add real-time voice AI to your apps, with the flexibility to integrate any AI model (custom or third party) for building voice agents.
My Role
Designed 6 / 8 dashboard pages end-to-end, co-created the Graph Editor with the lead designer, and led design QA to ensure visual and interaction consistency. Partnered with engineering to ensure seamless handoff and implementation.
Solution
Agora ConvoAI Studio: A voice agent management dashboard and an integrated graph editor (TEN Super Playground) for creating, editing, and deploying agent flows.
Includes support for template libraries, project dashboards, extension library and marketplace, flow editing with node-level configuration, metrics, and billing
Broader Adoption → A no-code/low-code interface opens conversational AI creation to non-technical users, expanding Agora’s reach.
Market Leadership → By launching an industry-first unified playground, Agora positions itself as the neutral hub for scalable, multimodal conversational AI.