AskPC — AI Chat Interface

Overview

AskPC is an AI-powered chat interface that lets analysts and investors query private market data using plain language. Before AskPC, users had to navigate PrivateCircle's Research product manually applying filters, browsing company pages and requesting data updates one by one. AskPC replaced that friction with a single question.

The Problem

After ChatGPT went mainstream, user behaviour shifted. People started expecting to just ask a question and get an answer. The Research product was powerful but required users to know where to look.

Old Way

Research → Apply filters → Browse company → Request update → Wait

New Way

Type your question → Get structured answer instantly

My Role

Designed the full product from scratch and built the frontend using Cursor. Went from first design to launch in 2 days.

How We Shipped in 2 Days

To move fast without cutting quality, we used Google Gemini's Material components as the base and applied PrivateCircle's design system theme, CSS and custom buttons on top. This meant we did not have to build every component from scratch. The foundation was solid, the product felt native to PrivateCircle, and we shipped without compromising on consistency.

The Hard Part

The biggest design challenge was not the chat interface itself. It was making users trust the answers. Private market data is sensitive. Analysts and investors make real decisions based on it. I used Zipy to watch real user sessions after launch and spoke directly with multiple users about trust. The pattern was clear users did not doubt the data as much as they doubted the process.

01

Real-time Progress Steps

Deep Research queries can take 5 to 10 minutes. Instead of a spinner, we showed exactly what was happening at each step Analyze Query, PC Search Agent, Create Plan, PC Company Agent, Web Search Agent. Users could see the work in real time. This was added post launch based on what we saw in Zipy sessions.

02

Structured Responses

Results return as formatted tables, financial comparisons and key insights sections. This made data feel organised and verified, not like a chatbot guessing.

03

Source Transparency

Every response shows where the data came from. Users can trace the answer back to the source, which is critical for financial decision making.

Result

2 Days

From design to launch

6,000+

Monthly active users

40+

Queries per session

Next Case Study →