AI Search
Senior Product Designer
2025
Background
The rise of AI
As we've all become aware, the rise of AI has fundamentally reshaped our expectations of digital experiences. Across the ecosystem, products have evolved from passive systems to intelligent, assistive tools, raising the bar for experiences that are faster, more intuitive, and better equipped to anticipate user needs.
Within this context, Affinity recognized the need to move beyond manual workflows to a more intelligent system aligned with evolving expectations. As a platform built to surface relationship insights, this meant enabling users to uncover insights and act on them more easily.
At the time, pressure was mounting from customers who felt Affinity was falling short of baseline expectations and began exploring more AI-forward alternatives. Within just a year, our survey showed a significant rise in customers planning to use AI in their daily workflows, highlighting how rapidly expectations had shifted.
While it was clear we needed to act, we wanted to do so thoughtfully, avoiding the temptation to build AI features without a clear understanding of their value.
👥 Team
I worked in a cross-functional squad that included 1 PM, 1 front-end engineer, 2 ML engineers, and a team of back-end engineers.
💼 Role
I was the sole designer and led the end-to-end design process from research and vision to final handoff.
Goals
Our goal
We set out to understand where AI could meaningfully improve the product, focusing on uncovering gaps in the current experience and identifying high-value use cases. While we knew customers were increasingly asking for AI and expected it to be part of the product, we lacked clarity on what that actually meant in practice. This required a research-driven approach to understand where and how it could deliver real value.
Our goal with this work was to define a clear direction for AI at Affinity, determine how it should fit into the product experience, and identify a starting point that could deliver meaningful value quickly.
Research
Growing pressure internally
To kick off the research, I partnered with my PM to conduct user interviews to better understand how customers were thinking about AI. To avoid bias, we kept conversations intentionally broad, focusing on uncovering gaps in the overall experience rather than framing questions through an AI lens.
After synthesizing insights, I developed a set of exploratory concepts and brought them back to users to evaluate and rank based on relevance to their day-to-day workflows. These concepts ranged from surfacing relevant connections for outreach to automating data extraction from emails and pitch decks.
Across multiple feedback sessions, the clearest need emerged: users wanted a faster, more intuitive way to ask semantic questions and get answers from their data in Affinity. This concept ranked #1 out of 6 with an average score of 4.3 out of 5 in perceived value.
While users could technically access this information at the time, doing so required a lot of manual effort, highlighting an opportunity for us to make information access easier and more aligned with evolving expectations.
Problem
Our problem
The core problem was that when users had questions about companies or people in Affinity, they relied heavily on keyword searches, filters, and manually digging through notes to find answers. In a fast-moving environment where investors are managing multiple companies and deals, speed is critical—making fast, frictionless access to information essential.
Search was a high-frequency behavior, but also a consistent point of frustration. While users relied on it to navigate companies, deals, and past interactions, it often fell short due to its rigidity and lack of flexibility. Despite some recent improvements, users increasingly expected a smarter search experience that aligned more with the capabilities of modern AI.
Assumptions
Hypothesis and assumptions
Based on customer demand, high usage, and what was feasible to deliver quickly, we identified search as the most effective starting point for introducing AI into the Affinity experience. As one of the most frequently used areas of the product, it was where users already spent the majority of their time, making it a natural place to meet them where they already are. From there, I partnered with my PM to define key hypotheses, assumptions, and opportunity areas.
Explorations
First explorations
After an initial pass at explorations on baking in search we realized insights that went against our hypothesis or we didn't account for. Customers didn't want it to be fully automated, but they wanted suggestions.
The larger ecosystem also needed to be developed as well on what a chat system looked like and where it would live in the navigation.
Although chat wasn't in scope for this project, I still wanted to account for what this looked like to help shape our road map. Early wireframes of chat…
Iterations
Iterations
After fleshing out concepts for semantic search we interviewed users to get feedback. We continued to learn and flesh out to make design decisions.
Final Designs
Final Designs
Semantic search (prototype animation)
Semantic search (prototype animation)
More detailed animation
Chat concepts
Impact
Impact
The core problem was that users had to rely on exact keyword searches, apply filters, and dig through notes, making even simple questions time-consuming.
Takeaways
Takeaways
The core problem was that users had to rely on exact keyword searches, apply filters, and dig through notes, making even simple questions time-consuming.