Friendli Suite
Redesigning B2B SaaS Product to Streamline Gen AI Workflows
The project redesigned Friendli Suite, a platform that streamlines generative AI model deployments. It simplified workflows, improved user experience, and clarified product identity to better serve AI engineers.
TIMELINE
Jan 2024 - Present
ROLE
Sole Designer
(User Flows, UI Design, Prototyping, Usability Testing, Visual Design and Design System)
FOR
WITH
4 Frontend engineers,
Product manager
ABOUT COMPANY
FriendliAI is a B2B company specializing in generative AI infrastructure solutions. Its mission is to empower organizations to harness the power of generative AI with ease and cost-efficiency.
ABOUT PRODUCT
Friendli Dedicated Endpoints offer a fully managed solution for deploying and serving generative AI models with dedicated GPU resources, streamlining fine-tuning, deployment, and infrastructure management to ensure high-performance, reliable, and cost-effective AI operations.
PROJECT OVERVIEW
Redesigning for Enhanced Usability and Cohesive Identity
In 2023, PeriFlow Cloud launched as a generative AI platform for training and serving models but faced usability challenges due to its complexity. Rebranded as Friendli Suite in 2024, it was fully redesigned to streamline complex workflows for AI engineers.
As the sole designer, I led the end-to-end transformation, simplifying workflows, enhancing information hierarchy, and creating a cohesive design system to deliver a more user-centric experience in the generative AI space.
Final Solution at a Glance
How does Friendli Dedicated Endpoints help AI engineers?
MAIN FEATURES
Endpoint
Create a team project
Fine-tuning
Create a fine-tuning job
Quickstart
Deploy a fine-tuned model
MAIN FLOW
01
Create a team project
02
Create a fine-tuning job
03
Deploy a fine-tuned model
04
Monitor and Test an endpoint
KEY FEATURES
Quick & Easy Model Deployment
With HuggingFace and Weights & Biases model repository integration, user can simply choose open-source model and deploy with few simple clicks.
KEY FEATURES
Streamline from
Fine-Tuning to Serving
Simplified the workflow by removing redundant steps like job comparisons and enabling direct integration with platforms like Hugging Face. Users can now fine-tune models and deploy them in a single step, cutting workflow complexity by 30-40%.
KEY FEATURES
Comprehensive Monitoring
Reorganized the dashboard to prioritize key metrics like latency, throughput, and logs, enabling faster issue diagnosis and proactive model optimization.
KEY FEATURES
Testing an Endpoint on Playground
Introduced a dedicated Playground tab for interactive endpoint testing with real-time feedback. Users can test queries and log results, making validation intuitive and boosting confidence in deployments.
Problem Context
What were the issues with PeriFlow Cloud?
BACKGROUND
A Redesign for Streamlined LLM Workflows
The redesign from PeriFlow Cloud to Friendli Suite focused on refining the product’s core features and identity. The original platform had too many features, making the user experience overwhelming.
PROBLEM CONTEXT
Feature Overload Without Focus
The platform offered a wide range of features, but many were not aligned with user needs, especially in the training workflow.
PROBLEM CONTEXT
Complex, Fragmented Workflows
Users had to navigate multiple steps to upload checkpoints, deploy models, and monitor performance, making the overall experience tedious and confusing.
PROBLEM CONTEXT
Weak Product Identity
I created a visually engaging Model List Page to show off the models we host, making it easy for first-time user to explore and quickly grasp our offerings.
Research for Addressing Issues
Why Redesign?
USER JOURNEY MAP
Design Criteria for the Friendli Playground
Based on the business goals, user challenges with existing playgrounds, and design objectives, I have established a set of guidelines to adhere to throughout the design of the Friendli Playground
USER RESEARCH
Design Criteria for the Friendli Playground
Based on the business goals, user challenges with existing playgrounds, and design objectives, I have established a set of guidelines to adhere to throughout the design of the Friendli Playground
04
Design Process
DESIGN CRITERIA
Key Redesign Goals
Based on the business goals, user challenges with existing playgrounds, and design objectives, I have established a set of guidelines to adhere to throughout the design of the Friendli Playground
Final Designs
IMPROVEMENT
Providing Pre-Deployment Information
We added pre-deployment context screens to clarify what happens when users deploy a model and what they can expect next.
BEFORE
AFTER
Final Solution at a Glance
Since January 2024, Friendli Playground has been continuously evolving by adding new features to keep pace with the rapidly changing trends in generative AI technology.
The Impact
Support Diverse B2B User Types
Most LLM playgrounds have the same problem—when errors happen during response generation, users are stuck with no clear options. I want to tackle this by giving users more control, like suggesting fixes or offering alternative workflows to keep things moving.
👈 Previous sketches of various model playgrounds!
Next Steps
Support Diverse B2B User Types
Most LLM playgrounds have the same problem—when errors happen during response generation, users are stuck with no clear options. I want to tackle this by giving users more control, like suggesting fixes or offering alternative workflows to keep things moving.
👈 Previous sketches of various model playgrounds!
08
Final Takeaways
Keeping Up with Rapid Tech Trends 🤯
Generative AI moves fast—new models claiming to be the best are released constantly, and groundbreaking technologies can feel outdated in just a few months. As a designer delivering these cutting-edge tools to users, I’ve learned the importance of staying on top of trends and paying attention to how users react. This experience has helped me grow as a designer working on AI products, teaching me how to balance innovation with usability in such a fast-paced space.
Communication Is Everything
In tech-heavy products, clear communication with developers and PMs is key. Aligning technical possibilities with user needs takes teamwork, and this experience has sharpened my ability to collaborate across teams to deliver both robust and user-focused solutions.
