- 2021/2022 -
Samba Suite for SambaNova Systems
Samba Suite for SambaNova Systems
Building the first Enterprise-grade generative AI platform
SambaNova Systems designs and creates Chips for ML. On top of it, they needed a product designer leader to build the software stack.
ROLE:
Team Lead -
Sr. Principal Product Designer
UI/UX
Product Strategy
User Research
Team Lead -
Sr. Principal Product Designer
UI/UX
Product Strategy
User Research
SCOPE:
14 Months
14 Months
PLATFORM:
Desktop
Desktop
TOOLS:
Figma
Adobe Illustrator
Figma
Adobe Illustrator
TEAM:
Six person
Six person
MARKET :
B2B SaaS
B2B SaaS
INDUSTRY:
AI/ML
AI/ML
INTRODUCTION
SambaNova is working to create an AI platform for businesses. The company targets enterprises that want to integrate sophisticated AI systems into their business yet lack the resources to develop and maintain in-house Data Science teams.
It was April 2021, and after raising $676M at a $5.1B valuation after four years of uninterrupted growth, SambaNova Systems decided it was time to expand its offer beyond hardware, creating its first software to run Machine Learning processes.
It was April 2021, and after raising $676M at a $5.1B valuation after four years of uninterrupted growth, SambaNova Systems decided it was time to expand its offer beyond hardware, creating its first software to run Machine Learning processes.
SambaNova's original offer was based on hardware deployment for Machine Learning.
THE PROBLEM
The company had a foundational hardware focus from its inception, and its teams didn't have software development experience. At that point, they decided to bring the first product designer to lead the effort to create DaaS (Data-flow-as-a-Service) software from scratch.
I was recruited in Q3 of 2021 and joined the company in October.
I was recruited in Q3 of 2021 and joined the company in October.
From the beginning, it was clear that to satisfy the company's potential users, I first needed to understand who they were, their needs, and their goals to create solutions for them. I was introduced to some preliminary drafts of the concept of building a platform that can run ML projects on the fly and with a no-code approach. It's important to say that the standard of the market is to do these experiments through CLI (command line interface) commands, so the users are not only Machine Learning and Data Scientist experts but also professionals who have advanced coding skills to develop their daily tasks using languages like Pytorch or R.
The challenge was to create a platform that could replicate traditional ML flows and processes without a single line of code, expanding the number of potential users from thousands to millions.
The platform's must-have feature is the capability to run Machine Learning Computer Vision and Natural Language Process (NLP) jobs without having deep experience in the space.
The challenge from a UX perspective was replicating extremely complex code-based tasks most efficiently through an intuitive GUI.
The challenge from a UX perspective was replicating extremely complex code-based tasks most efficiently through an intuitive GUI.
RESEARCH
I started a series of qualitative research studies with internal stakeholders to understand the market and users' needs. The ultimate goal was to identify User Personas, their goals, and pain points to define the must-have features for an initial release, building a product for them.
To understand the different layers of information and how they are interconnected I created a Product Design Strategy Map, where every voice and insight is located based on its relevance and relationships.
The research was split into stages using different methods targeting people in critical positions on these layers, from a user-operational and also business standpoint. I embarked on the task of validating and aligning every stakeholder under the same vision from the company and product go-to-market strategies to the target, market, verticals, and final users of the product understanding how each layer of information should impact the product itself and at what level.
To understand the different layers of information and how they are interconnected I created a Product Design Strategy Map, where every voice and insight is located based on its relevance and relationships.
The research was split into stages using different methods targeting people in critical positions on these layers, from a user-operational and also business standpoint. I embarked on the task of validating and aligning every stakeholder under the same vision from the company and product go-to-market strategies to the target, market, verticals, and final users of the product understanding how each layer of information should impact the product itself and at what level.
Interviews and research methodologies
• Card sorting for flow definition
• A/B testing for fonts and light-dark mode definition
• Usability tests for UI/UX validation
• Walkthroughs for intuitiveness validation and feedback gathering
• UX/UI competitive analysis
• Visual cultural references (gaming - sci-fi films)
• Interviews for:
- market/verticals requirements
- company go-to-market strategy
- product go-to-market strategy
- company's external messaging
- company's branding concept and visual definition
- external messaging
- persona definition
- insights detection
• A/B testing for fonts and light-dark mode definition
• Usability tests for UI/UX validation
• Walkthroughs for intuitiveness validation and feedback gathering
• UX/UI competitive analysis
• Visual cultural references (gaming - sci-fi films)
• Interviews for:
- market/verticals requirements
- company go-to-market strategy
- product go-to-market strategy
- company's external messaging
- company's branding concept and visual definition
- external messaging
- persona definition
- insights detection
Some of the questions for the Usability Test Survey
1- What is your role?
2- What is your background and experience?
3- How many years of experience in Machine Learning do you have?
4- Is this the first time you are using the product? How much training do you have to use it?
5- Do you think the product follows certain branding guidelines?
6- Is it clear to you the company alignment in terms of look and feel?
7- Do you think it should follow specific principles? Which ones?
8- Do you think it is hard to use? 1-5 rank
9- Do you think the interface is intuitive? 1-5 rank
10- Rank UI, UX, intuitiveness, flow, color use, information architecture
11- Choose the best words to describe the User Interface:
Clean - overloaded - intuitive - complex
12- Do you think a new customer/user would need a lot of training to use it?
13- Do you have any suggestions or missing elements?
2- What is your background and experience?
3- How many years of experience in Machine Learning do you have?
4- Is this the first time you are using the product? How much training do you have to use it?
5- Do you think the product follows certain branding guidelines?
6- Is it clear to you the company alignment in terms of look and feel?
7- Do you think it should follow specific principles? Which ones?
8- Do you think it is hard to use? 1-5 rank
9- Do you think the interface is intuitive? 1-5 rank
10- Rank UI, UX, intuitiveness, flow, color use, information architecture
11- Choose the best words to describe the User Interface:
Clean - overloaded - intuitive - complex
12- Do you think a new customer/user would need a lot of training to use it?
13- Do you have any suggestions or missing elements?
Color low-fi A/B testing, Competitive analysis, and film GUI references
THE TYPEFACE SELECTION
The initial idea was to use the company fonts used for branding and marketing assets: Century Gothic and Avant-Garde. The alignment with branding was to add "fresh air" to SambaNova's look and feel through its flagship product, and these fonts are very 80s and 90s style. Also, they are not the best option for screen-digital use as they were created many years ago, even before monitors even existed.
The goal was to find a unique font that could be used through Google fonts license, screen-first, and with. a monospace variable as a way to bring a "technical-computing-style" flavor to display data.
To find the best options and present them to the Branding and Marketing team, I created a typeface report to evaluate and compare possible options for the Product.
To find the best options and present them to the Branding and Marketing team, I created a typeface report to evaluate and compare possible options for the Product.
After carefully reviewing and evaluating the pros and cons, the decision was to move forward with Roboto for the number of available variables, performance, readability in small sizes, and the option to use Roboto Mono for data visualization.
UNDERSTANDING THE USERS
I defined experience maps, low-fidelity artifacts, and the initial workflow, which data scientists in card-sorting research validated to confirm their working methods.
Whiteboarding session to define the workflow.
After gathering multiple stakeholder insights, different personas were identified, also enterprise readiness and requirements for different industries and verticals. I defined AI/ML alignment in terms of wording and best practices.
THE PROCESS
When I joined SambaNova Systems they had already some very early-stage flows, UI and UX. Even after the research, the findings validated some of the assumptions, the flows, and the Information Architecture required a restructuration.
Early stage wireframes and flows I received when joining the company.
From the interviews with the C-Suite, I learned there were going to be other products and projects that would require UI. With that in mind, I decided to create a Design System with Master Components to standardize screen development to have cohesiveness between products and to speed up the process. The Company GTM (go-to-market) strategy in terms of marketing was to project a unique look and feel for every asset and my task was to balance that visual style at the product level incorporating the company's visual style into the products. The critical part was to define which elements from the Branding Guidelines were able to be applied and which ones required a product design approach, like fonts and color palette application.
This Design System grew from basic elements to a complex library of components and rules for building software. From color applications to different hierarchies of buttons, inputs, dropdowns, and selectors, to responsive breakpoints and rules for creating screens and flows; I put together all kinds of scenarios and options for junior designers and front-end engineers. Built seamless experiences almost without any supervision and just a final design QA review.
A couple of pages from SambaNova's Design System. They are organized by categories like Color, Typefaces, Buttons, Interactive elements, Inputs, Menus, Selections, Tables, Cards, etc.
THE SOLUTION
Understanding the pain points and needs of the users was the first step to creating a solution where not-very-skilled operators can run Machine Learning processes most intuitively.
As speed was an important aspect to consider I started with Material Design components as a foundation of SambaNova's User Interface elements. After aligning Branding, Marketing, and Product behind a cohesive and unified experience across every channel I created an extremely clean seamless, delightful, and beautiful experience.
The product was tested and validated using interactive prototypes I created in Figma by Data Scientists and Machine Learning. After just a 2-minute explanation about the Information Architecture of the platform, the participants of these tests were able to run ML jobs without any kind of previous training achieving an exponential reduction in their workload and time assignment for achieving the same tasks using CLI (command line interface).
The result is a very clean and intuitive no-code platform,
where users can execute the most complex ML work.
where users can execute the most complex ML work.
SambaNova Suite release video - 02/2023
Team members
Azadeh Riahi
Product Manager
Product Manager
Shuya Zhan
Product Designer
Product Designer
Rajesh Ottikunta
Front End - Software Engineer
Front End - Software Engineer
Kenneth Huang
Front End - Software Engineer
Front End - Software Engineer
Bill Bain
Content Strategist
Content Strategist
