Agents
AI research agents, tailored to you.
Tasks
One task in. One finding out.
Queue a task — re-underwrite on the next 10-Q, break down segment margins, check a competitor's capex. The agent executes and reports back with something structured, not a wall of text.
Some tasks run once. Some run forever — like watching an IR page every week and pinging you only when something changes.
Research tasks
3 agents activeRe-underwrite on Q1 10-Q
triggered by filing
Segment-margin breakdown
one-time
Read 10-Q, draft thesis update
62% · ~8m left
Watch competitor IR page
weekly
Datacenter-mix finding
conviction +6
Capex cross-check
no change
Monitoring
Your sources, watched continuously.
Point agents at the filings, transcripts, Substacks, X accounts, and RSS feeds that matter to your positions. They index the source documents and surface what's relevant.
You get the signal without doing the daily sweep yourself.
Source documents · NVDA
Q1 FY26 Quarterly Report
May 2026
Q1 FY26 Earnings Transcript
May 2026
FY25 Annual Report
Feb 2026
Investor Day — slide deck
Jan 2026
Q1 FY26 · pulled from 10-Q
Insights
See reality test your thesis.
As evidence lands, agents mark it against what you believe: confirmed, contradicted, or missed. Each insight is dated and placed on your timeline.
Track speculative claims as hypotheses — and accept them into the thesis only once the evidence accumulates.
Research timeline · NVDA
Pan left to see the assumptions, evidence, and decisions that built conviction. Pan right to compare those calls against realized outcomes.
Keep the judgment. Delegate the legwork.
You decide what to track, what to investigate, and what to believe. The agent handles the rest. Request access to the private beta.
Private beta · by invitation