AI-powered investment research, from discovery to conviction.
Cerebellum brings company research, evidence evaluation, thesis development, and portfolio risk analysis into one connected workspace. Investigate companies with sourced data, challenge your assumptions with AI-assisted audits, and monitor whether your reasoning still holds as the evidence changes.
Move from an idea to a company worth investigating.
Cerebellum gives you several entry points into research. Start with a screener to find candidates, track tickers you're watching, compare companies directly, or explore macro and sector context to understand the environment.
Stock Screener
Run a multi-dimensional analysis on any ticker. Start from a name, a sector, or a community post and go straight into a full research breakdown.
Open ScreenerWatchlists
Add tickers to a watchlist and Cerebellum monitors them for material developments. The watchlist also feeds your daily briefing and research context.
Manage WatchlistCompany Comparisons
Stack two companies side by side across fundamentals, valuation, performance, and intelligence scores to evaluate which better fits your thesis.
Compare StocksMarket & Sector Context
Global market indices, macro indicators, and economic calendars help you contextualize company research within the broader environment.
View Global MarketsBring company information together for more informed research.
Cerebellum pulls together the data investors need and interprets it, not just displays it. Sourced facts, calculated metrics, and AI-generated interpretations are clearly separated so you know what is verified and what is an estimate.
Financial Fundamentals
- Revenue, margins, EPS, and growth rates from filed financials
- Balance sheet: debt, cash, equity, and return on equity
- Free cash flow and capital allocation
Valuation Models
- DCF, EV/EBITDA, P/E, PEG, and reverse DCF fair values
- Metrics calculated deterministically in code, not by AI
- Every user sees the same numbers for the same ticker
SEC Filings & Disclosures
- 10-K, 10-Q, Form 4 insider transactions, 13F holdings
- Earnings call summaries and key quotes
- Source links to original EDGAR documents
Analyst & Institutional Activity
- Analyst consensus and price targets (external, not recommendations)
- Insider buying and selling patterns
- Institutional ownership shifts from 13F filings
Technicals & Sentiment
- RSI, MACD, moving averages, and Bollinger Bands
- News sentiment and social signal analysis
- Chart visualization via TradingView widgets
Sources & Evidence
- Sourced facts, calculated metrics, and AI interpretations are clearly separated
- Source links on material claims for independent verification
- Data status indicators show whether a metric is verified, estimated, or missing
Data sources include SEC EDGAR, Financial Modeling Prep, analyst consensus feeds, and macroeconomic indicators. Market data is end-of-day or delayed, not real-time. International fundamental data relies on AI web search and may be less complete than US-listed data. AI-generated interpretations do not replace original filings or verified financial data.
Convictions Lab: your reasoning becomes a persistent asset.
The Thesis Lab is a core differentiator, not a note-taking feature. It structures your investment reasoning and then challenges it, so you can evaluate your logic with evidence rather than emotion.
Record your thesis
Document why you own a stock before you buy: bull case, bear case, catalysts, risks, and invalidation conditions. The structure keeps your reasoning intact when the price moves but the facts haven't.
Identify catalysts and assumptions
Make the assumptions supporting your thesis explicit. Each one becomes a checkpoint the system can monitor for evidence that supports or contradicts it.
Establish risks and invalidation conditions
Define what would prove your thesis wrong before you're emotionally invested. This makes it harder to move the goalposts when the price drops.
Challenge with AI-assisted audits
The Adversarial Audit searches for evidence that contradicts your assumptions, identifies cracks and contradictions, and flags what would prove you wrong. Cognitive Analysis looks for reasoning patterns like confirmation bias and anchoring across your thesis versions.
Revisit as evidence changes
The Thesis Health Score tells you when the evidence has shifted, not just when the price moved. Every thesis version is saved so you can see how your reasoning evolved.
The purpose is to encourage disciplined reasoning, not to tell you what to buy or sell. The system does not identify every cognitive bias or determine whether an investment will succeed.
An AI research agent that shows its work.
Bella classifies your intent, selects tools from a registry, plans multi-step research, and synthesizes results into a structured format. She is transparent: you can see which tools ran and why. She pulls real data through the tools and interprets it, rather than fabricating answers.
Some actions still require user input. Bella is a research assistant, not an autonomous decision-maker. She can be wrong, incomplete, or biased. Every finding should be independently verified against source evidence.
Cerebral Core: cognitive tools for disciplined reasoning
Cerebral Core is a set of cognitive tools available to Premium subscribers that structure your research memory, map relationships, and profile your investing patterns.
Keep up with the information that may affect your research.
Cerebellum connects new information to the company or thesis it affects, so you don't have to manually check every position every day. Alerts are triggered by material changes, not by every market tick.
News & Article Discovery
Cerebellum surfaces news and developments relevant to the tickers and sectors you follow, with source links for verification.
Filings & Disclosures
SEC filings, earnings releases, and insider transactions are tracked and connected to the relevant company or thesis.
Notifications & Research Updates
You receive alerts when something material changes on your watchlist, not on every market tick. Alerts link to the relevant article, filing, or analysis.
Thesis Health Monitoring
The Thesis Health Score tracks whether the evidence supporting your conviction has shifted. It tells you when to revisit, not just when the price moved.
Cerebellum uses end-of-day and SEC filing data, not real-time trading feeds. It does not promise continuous surveillance, comprehensive news coverage, or automatic detection of every material event. Supply-chain developments and institutional activity are tracked where data is available.
Understand your portfolio, not just watch tickers.
Portfolio-level analytics help you answer practical questions: whether your portfolio is concentrated in a particular sector, how it behaves relative to a benchmark, and what certain market scenarios could mean for your holdings.
Holdings & Allocation
Manage holdings with target allocation weights and see how capital is distributed across positions.
Concentration Risk
See how much capital sits in a single name, sector, or factor, and whether your portfolio is over-exposed.
Volatility & Drawdown
Understand how your portfolio behaves under stress, with historical volatility and drawdown metrics.
Factor Exposure
Momentum, value, quality, and volatility signals show what drives your portfolio's returns.
Correlation Clustering
Discover hidden overlap between positions that may look different but move together.
Scenario Analysis & Stress Tests
Model rate hikes, recessions, supply chain shocks, and inflation scenarios against your actual holdings.
Stress-test results are modeled scenarios, not predictions. Historical metrics are backward-looking and do not guarantee future performance. Portfolio risk analytics are available to Pro subscribers. See the pricing page for current plan details.
Discuss ideas and exchange research with other investors.
Cerebellum's community is a complementary part of the platform. It exists alongside the research tools, not in place of them. Investors can share analysis, discuss ideas, and learn from each other.
Community opinions do not constitute verified research or professional financial advice. Every investment decision is yours and should be based on your own due diligence.
Responsible AI and verifiable research.
Cerebellum separates verified data from estimates and AI interpretation. Source attribution, data status indicators, and ownership enforcement are built into the platform. These are the controls verified in the running application, not aspirational claims.
Server-side ownership enforcement
Every backend function verifies record ownership before returning data. Queries use system-enforced created_by_id, not client-settable fields that could be spoofed.
Row-level security on private data
Portfolios, theses, watchlists, messages, and alerts are isolated at the database level. Users can only read and modify their own records.
Prompt injection defense
External content (SEC filings, web pages, user input) is scanned for injection patterns and wrapped in data boundaries before entering AI prompts.
Deterministic metrics, not AI guesses
Financial metrics are calculated in code, not generated by language models. Every user sees the same P/E, EV/EBITDA, and DCF values for the same ticker.
API key isolation
User-provided AI API keys (BYOK) are stored securely and never exposed to the frontend. The external Agent API only accesses public company data, never private user records.
Post-LLM validation
After the AI responds, programmatic checks validate source URLs, date freshness, and financial figure accuracy against computed metrics. Impossible values are nullified.
Cerebellum uses platform-provided authentication and infrastructure. Application-level controls include ownership enforcement, RLS, prompt injection defense, and post-LLM validation. This is not a claim of end-to-end encryption, zero-knowledge architecture, or independent security certification.
A connected research process, not a collection of tools.
Cerebellum's value is not in any single tool but in how they work together. Research feeds your thesis. Your thesis drives your portfolio. Your portfolio powers alerts that tell you when to revisit. The context flows between stages without losing meaning.
Start your research workflow.
Research a company, build a thesis, and let Cerebellum monitor whether your reasoning still holds as the evidence changes. Free to start.
Disclaimer: Research insights on Cerebellum are synthesized from publicly available financial information for informational and research purposes only. This is not investment advice. Every investment carries risk, including the possible loss of principal. Past performance is not indicative of future results. Always do your own research and consult a qualified financial advisor before making any investment decisions.