September 7, 2026AI
Vega matches research depth to the question
Choose Quick for a lookup, Standard for comparisons, Deep for a research memo, or Auto to let Vega route the question.
- The credit hold is quoted before you submit
- Unused credits are returned, and the chosen depth stays on the answer
- A short balance keeps the prompt and offers a cheaper route
- Comparisons cover two to eight assets in one frame
September 5, 2026AI
Vega reports preserve the evidence trail
A Vega thread can now become a structured PDF report with its tables, charts, analytics, backtest measures, headlines and cited sources.
- A key-figures summary and numbered sections for long threads
- Dataset references keep as-of dates, methodology notes and caveats
- Public PDFs are built from the sanitized shared page, without owner-only context
- Public reports show evidence by default, and owners can add Vega's discussion
August 31, 2026AI
Vega charts explain and export their evidence
Every Vega chart now has a factual caption derived from the plotted series, plus the data needed to reproduce it.
- Download any chart's plotted data as CSV, including older conversations
- Export a thread's charts to one Excel workbook or a PDF
- Captions name the series, date range, direction and material moves
- Shared research keeps charts and CSV exports without owner-only links
August 29, 2026Data
A full-universe equity screener
Build a screen across the US equity universe, choose the columns you want and keep the result reproducible in the URL.
- Filter valuation, profitability, growth, performance, volatility, size, liquidity and dividends
- Narrow by index, sector, industry or exchange, with price and dollar-volume floors
- Up to four sort keys, the full match count and CSV export
- Presets, and a URL that saves the screen so you can revisit or share it
August 23, 2026Analytics
Aegis turns portfolio changes into auditable decisions
Model a target book, compare it with current exposures, test both against named limits, and keep the evidence as holdings change.
- Decision Review classifies each change and shows modeled-data coverage per book
- Policies test VaR, CVaR, volatility, exposure, beta, concentration, position size, turnover and coverage
- Monitoring records each evaluation and alerts when a check moves into breach
- Saved records freeze the books, policy revision, provenance and verdict, with private JSON and CSV exports
August 22, 2026Analytics
Aegis follows the portfolio that actually existed
Track Record now rebuilds performance from recorded history instead of applying today's holdings to past markets, and stops when the lineage is incomplete.
- Accepts a validated run, a paper deployment's model NAV or a saved portfolio
- Paper histories are labeled model NAV, simulated and not broker-verified
- Portfolio composites use effective-dated allocations and stop at missing sleeve history
- Each record carries its return basis, approximation state and data coverage
August 3, 2026Analytics
Portfolio risk answered the same way everywhere
Every surface that asks for a book’s risk now gets the same answer from the same computation.
- Sleeves net once on the server, for the portfolio view and Aegis
- The Aegis deep link no longer rounds away small positions
- A strategy target the preview cannot reproduce is refused instead of shown
- Paper performance is reported and labeled on a model-NAV basis
August 1, 2026Platform
Strategy exports you can pin and download
Exports now pin to a strategy version, include a receipt and download from the browser, the command line or MCP.
- Every export carries its strategy version and a content hash
- Pine Script export refuses strategies it cannot render faithfully
- Export eligibility comes from running the exporter, so the badge matches the file
- Exporting a strategy says nothing about broker eligibility
- No broker connection is live, and nothing here places an order
August 1, 2026AI
Vega covers global indices and names its limits
Comparisons now reach the major global indices, and when Vega hits the edge of what it supports, it says which limit it hit.
- Global index comparisons across SPX, FTSE, DAX, Nikkei and Nifty
- A named dataset resolves to that dataset instead of a nearby one
- A declined question says whether it hit a coverage gap or an unsupported request
- Figures the data cannot support are refused rather than estimated
- News evidence marks absence explicitly, so a quiet week reads as quiet
July 29, 2026AI
Vega reads your saved portfolios
Ask Vega about a saved portfolio and it reads the real book, with named portfolios going straight to the risk view.
- Saved portfolios from Vega and Aegis are readable by name
- Factor risk and the top ten contributors use the cockpit's computation
- Charts render on the server at full resolution, so long windows are not thinned
July 27, 2026Analytics
Aegis reports you can hand to a client
The risk cockpit now exports a structured CSV and a detailed vector PDF in place of the old raw CSV dump.
- Vector PDF report with TradePolaris and Aegis branding
- Sections on methodology, confidence, concentration, factors, correlation, drawdown, liquidity, options and caveats
- Both download straight from the cockpit
July 27, 2026Platform
Deployment reports and signals that expire
Generate a report on any deployment on demand, and every signal now carries an explicit expiry.
- Expiry uses market-aware arithmetic for sessions, weekends and holidays
- Expired signals are dropped at delivery, and expiry errs early, never late
- A guided first deployment, and a signal feed you can copy or export as CSV
- Deployments remain simulated, with no broker connection and no orders
July 19, 2026Analytics
Portfolios now hold options and futures
A book can now carry option legs and futures positions alongside equities, and both flow through to risk.
- Option legs on sleeves are folded into live risk
- Futures positions use the correct CME contract multipliers
- Describe holdings in plain English on import, or re-import into an existing portfolio
- Option legs carry through to the Aegis cockpit and Vega chat
- Factor-model covariance gives several-hundred-name books a real risk model
July 19, 2026Platform
A strategies library and draft recovery
Strategies now live in one library, and the builder keeps a draft you were partway through.
- A stepper for starting a new strategy
- Draft recovery, so a half-finished strategy survives a closed tab
- The builder reveals options step by step instead of all at once
- A full walk-forward run asks you to confirm credits first
July 18, 2026AI
Saved apps on the dashboard, refreshed overnight
Saved apps now appear on your dashboard and on one board, and re-run after the close so they are current each morning.
- An unfinished draft is restored in the builder at no cost
- A board shows every saved app's snapshot on one grid
- The worker refreshes apps nightly after the close, not on your next visit
July 18, 2026Data
Quote updates that push, and charts that remember where you were
Tier-1 quote updates now push through a conflated fan-out instead of polling, with a watchdog on feed freshness. The feed is delayed by at least 15 minutes.
- Professional TradingView charts remember chart state per account
- The asset page keeps your place when the chart loads, and first bars paint sooner
July 15, 2026Data
More of the market in the data lake
New datasets cover US short interest, factor research and Indian derivatives.
- FINRA short interest and short volume, IPOs, and deeper minute-level equities data
- A Ken French factor dataset, refreshed weekly
- NSE futures and options across 159 names, plus 18 global and NSE index datasets
July 13, 2026AI
Vega Apps build dashboards from a sentence
Describe a chart or monitor in plain English and Vega builds an app you can save, refresh, edit and share.
- Clone gallery templates and rebind them to your symbols at no credit cost
- Refine in conversation, so “make it weekly” edits the same app
- “Watch this” turns a single-symbol app’s condition into an Updates inbox alert
- Public share links let anyone view an app and fork it
July 13, 2026Platform
New Pro, Aegis, and Funds & RIAs plans
Pro ($49/mo) is for recurring research, Aegis ($249/mo) adds the institutional risk suite with a 5,000-credit monthly grant, and Funds & RIAs start at $1,000/mo.
- Pro: 1,000 credits/mo, 10 paper-trading slots and 10 saved Vega apps
- Aegis: the full risk cockpit, everything in Pro and 20 saved Vega apps
- Funds & RIAs: multi-book portfolios, onboarding, book migration and LP-ready reporting
- Charts and market data stay free, and free accounts get 200 credits monthly
July 12, 2026Analytics
The Aegis institutional risk cockpit
A tabbed cockpit puts ex-ante VaR, stress testing, factor risk and the optimizer one click apart, recomputed as you edit the book.
- Ex-ante VaR, CVaR, factor decomposition, exposure and concentration that update with the book
- Stress testing with crisis replays, macro and factor shocks, and Monte Carlo
- An optimizer with sector and turnover constraints, Black-Litterman views and a never-worse minimum-variance mandate
- Multi-strategy sleeve portfolios with aggregate risk
- Deep links to every view and plain-English risk narratives
July 2026Data
Global search covers 13F filers, politicians and activists
⌘K search now covers thousands of managers in the recent 13F filing directory, with accession-level positions loaded on demand.
- Jump from any indexed 13F filer to its latest filing
- Politicians and 13D/G stake filers resolve by name in the same bar
June 2026AI
Vega, an AI research analyst
Vega queries the data lake, searches the web, drafts a strategy and runs a preview backtest, showing each step with its SQL and sources.
- Builds charts and tables from a plain-English request
- Citation cards pair each answer with a source, a chart and the query
- Proposes full validation or paper deployment, then waits for your confirmation
- @-mention any stock or dataset to pull it into the conversation
- A built-in cost meter keeps spend visible
June 2026Analytics
Holdings analytics against the S&P 500
Backtest reports now chart the equity curve against the S&P 500 and break performance down by holdings.
- Benchmark overlay with alpha, beta, R², tracking error and information ratio
- Up/down capture and a Fama–French 5-factor plus Momentum decomposition
June 2026Data
Live news feed and latest headlines
A live news feed with per-ticker sentiment now runs across the terminal, and you can ask the data for a name's latest headlines.
- Per-ticker sentiment with an AI-scored overlay
- Inline on asset pages and queryable in plain English
June 2026Workflow
Conditional alerts and an Updates inbox
Set conditional alerts on the data you care about and receive them, along with strategy events, in an in-app Updates inbox.
June 2026Engine
Honest, survivorship-aware backtests
The engine uses historical membership where available, keeps later-delisted names where supplied, warns when a run falls back to current constituents and models realistic trading costs. Backtested results are hypothetical and not indicative of future results.
- Delisted-symbol universe option on the full tier, with universe-mode warnings
- Next-bar-open fills, gap-aware stops, bid/ask spread, square-root impact and borrow costs
- Walk-forward, out-of-sample validation instead of one in-sample curve fit
June 2026Data
Index and sector screens
Screen by index membership (S&P 500, Nasdaq-100, Dow) and by sector with curated, point-in-time data.
- Real index levels and ~3 years of price history across 11 indices
- Sector heatmap, breadth and most-active names on the markets dashboard
June 2026Analytics
Tearsheets and on-demand analytics
Every backtest produces a tearsheet, and the same metrics run on any ticker, basket or benchmark without a backtest.
- Performance: total return, CAGR, Sharpe, Sortino, Calmar, monthly-returns heatmap
- Risk: volatility, max drawdown, longest underwater stretch, Ulcer index, VaR and CVaR (95%/99%)
- Overfitting: Probabilistic and Deflated Sharpe, Minimum Track Record Length, Probability of Backtest Overfitting
June 2026AI
Deep Research mode for Ask the Data
Vega now resolves universes it can't list directly, like the Fortune 500, with web search, then pulls the figures from licensed data instead of scraped estimates.
June 2026Platform
Usage-based credits and shareable tearsheets
Pay only for the compute you run, and share any backtest report through a public link with a social preview card.
- 100 credits cost US$1, with 200 free credits monthly and no card
- Ask the Data and preview backtests are free
- A full walk-forward backtest starts at about 96 credits ($0.96) and scales with history, factors and positions
- A referral waitlist with line-jumping
TradePolaris is research software, not investment advice. Backtested and hypothetical results do not predict future performance. Trading is paper (simulated) only. See our Disclaimer.