Personification

Inspiration

The idea came from a colleague whose well-designed app was rejected not for quality, but because it didn’t match how the decision-maker thought or preferred things. We realized design often fails due to misaligned understanding, not execution.

What it does

Personification turns real people and real design briefs into actionable personas. It analyzes authored text or public signals, aligns them with a design problem, and delivers clear guidance plus a short spoken briefing designers can listen to before building.

How we built it

We use Gemini on Vertex AI to analyze writing style and design context, synthesize a focused persona, and generate clear design recommendations. ElevenLabs converts that insight into a natural voice briefing, making the persona fast and easy to absorb.

Challenges we ran into

Public profile data is unreliable and often blocked, so we shifted to analyzing authentic writing samples instead. Balancing useful inference without over-speculating was also a constant challenge.

Accomplishments that we're proud of

We built a working end-to-end MVP that produces practical, context-aware personas and delivers them through voice not just text making the output immediately useful in real design workflows.

What we learned

Personas are far more effective when tied to a real person and a real problem. Voice summaries also help teams internalize insights faster than traditional documents.

What's next for Personification

We plan to support more data sources, improve confidence scoring, personalize voice styles, and expand beyond design into content strategy, product decisions, and stakeholder alignment.

Built With

• gcp
• nestjs
• next
• vertexai