Arcane
AI Intelligence Agent for Decision-Makers
Built in one week at CMU's Prodhacks.
Summary
Scans the internet for you. Delivers one high-signal update, daily.
The Prodhacks build that grew into Bulletyn.
Context
Solo, end to end, at CMU's Prodhacks.
Research · Framing · Strategy · MVP
The Problem
Interviews put numbers on the fatigue.
Arcane · user interviews · Prodhacks 2026
What I Built
Shipped inside the hackathon week.
Product Approach
One update per day, by design.
Feeds→Agent
Clicks→Clarity
Volume→Signal
Outcomes
Research & Validation
What interviews surfaced
The pain isn't reading. The pain is filtering.
Not more content. Three things:
Another feed to keep up with.
→One reliable daily digest, delivered.
Arcane · user interviews · Prodhacks 2026
User Journey
The broken loop before Arcane
Intent
Wants industry signal.
Hunt
Bounces across platforms.
Fatigue
Filters noise by hand.
Arcane compresses this loop into one delivery.
System Architecture
High-Level Workflow
1. Input & Pipeline
2. Filtering & Synthesis
3. Impact & Delivery
Market & Strategy
Value creation over raw engagement.
Time & Confidence
Ad-Supported Growth
Premium Subscription
B2B Expansion
Constraint increases trust.
One update a day earned more trust than five. Signal prioritization is the moat.