#AI
#B2B
AI-POWERED DATABASE
(Apprenticeship, Launched in the internal platform)
I designed an AI-powered SQL IDE, from concept to launch, that enables both technical and non-technical users to rewrite queries for faster database performance. The project was featured in an internal Google Lunch & Learn, showcasing its potential to improve enterprise-scale data workflows.
Business AI
Concept to Creation(0-to-1)
Sep 2024 - Dec 2024
Product Designer
Frontend Developer
Why not just Chat-GPT?
ChatGPT can handle simple queries, but enterprise SQL is far more complex, with nested subqueries, multi-table joins, and business-specific logic. Our IDE is built to understand these structures and deliver reliable, optimized queries that generic AI tools cannot.
How our IDE is different
I designed an AI-powered SQL IDE that goes beyond text generation. Unlike ChatGPT, it provides:
Context-aware query rewriting with enterprise-scale logic in mind
Performance optimization suggestions surfaced directly in the IDE
Seamless workflow integration so users can edit, run, and validate queries in one place
Quick overview of this project timeline
How I work with the team
I followed a 1-week Agile sprint cycle, using JIRA to manage tasks and track progress. Updates and blockers were shared in real-time via Slack, and we continuously refined our approach based on feedback and expert insights.
JIRA Board
Daily Stand-up
Retrospective with Mentor
Why an AI-powered SQL IDE?
Through user interviews with both SQL experts (like data scientists and software engineers) and non-SQL users (such as marketing analysts), we uncovered three recurring challenges:
1
The Struggle of Writing Efficient Queries
Marketing analysts often struggle when large datasets cause queries to fail, and ChatGPT can’t help because it doesn’t know the database schema
2
Time Wasted on debugging performance
Database administrators spend hours debugging slow queries without clear AI-driven performance insights
3
Lack of consistency in query standards
Software engineers face inconsistent query styles across teams, making debugging and maintenance harder
From scattered pain points to a clear persona that guided our design decisions
Bridging the Gap, Turning Ideas into Action
With user needs defined, I quickly explored solutions through a low-fidelity prototype, mapping out layouts and interactions for early validation. This rapid iteration helped align stakeholders before refining the design in Figma, where I built the first high-fidelity prototype, ensuring usability and AI-driven functionality were seamlessly integrated
Sketch
Low-Fidelity Prototype
Explain how do you think? And why you choose the upcoming ideas to go forward
In our Agile retro, I gathered feedback from our Google mentor and team to identify improvement areas.
❤️ Likes
1
AI-powered query optimization intrigued users
2
Side-by-side query comparison made it easier to evaluate improvements
3
Visual runtime difference indicators helped users grasp performance gains
😍 Wants
1
Clearer button naming for better action clarity
2
More transparency in AI-generated outputs
3
Improved execution control to prevent redundant query submissions
1. Users now see their input on the left and AI suggestions on the right, making comparisons effortless
2. Added performance metrics to show AI-improved efficiency at a glance
Before
One-Side Query
After
Side-by-Side Query
Before
Normal CTA text
After
Renamed "Optimize with AI" to
"AI Rewritten" for better clarity
New Features - Users can select from common query patterns, reducing the need to start from scratch
Smart Query Input
Users can write and run their own SQL queries, with built-in safeguards that prevent AI rewriting before any input is provided.
AI-Powered Rewriting
With one click, the IDE generates an optimized version of the query and highlights key differences side-by-side
When our frontend engineer Zhiqian had an accident. We quickly adapted by redistributing his unfinished tasks across the team to stay on track.
Solve technical problem
I solve the problem with my frontend experience ✨
My frontend experience allowed me to step in during team challenges, ensuring continuity and a smooth user experience.
Success Metrics
In 4 months
designed, developed, and launched from 0 to 1
86%
reduction in query time
(TPC-H: 255.77s → 35.92s,
SSB: 22.7s → 2s)
98%
user satisfaction, validated through 5+ in-depth user interviews
What I learned
✅
Beyond UI/UX, effective product design means understanding user pain points, business goals, and technical feasibility to create impactful, scalable solutions
✅
Building AI-driven features requires balancing automation and user control. Users trust AI when given clear explanations and the ability to refine results
✅
Working closely with engineers, researchers, and mentors reinforced the importance of iterative feedback and agile workflows for refining solutions
