A personal CIO · Private beta

The investment discipline of a private bank. Without the private bank.

Portivo is a personal CIO for serious self-directed investors. You write your investment policy once. Every scan, every research memo, every thesis, every sell — checked against it, logged, and saved as your knowledge compounds.

Your Investment Policy Statement, appliedYour Knowledge Base, compounding

Software for self-directed investors. Not investment advice, not personalized recommendations. You operate it; you decide.

Is this for you?

Portivo is a power tool. It is built for a specific kind of investor.

This is not a robo-advisor and not a chat-with-a-friendly-AI app. It will feel like overkill if you don’t recognize yourself below — and like the thing you’ve been trying to build in spreadsheets if you do.

If any of these sound like you, keep reading.

  • You manage $250K–$10M of your own money. Not a hobby account — material capital you actually rely on.
  • You think in policy, not vibes. Position sizing, asset class caps, cash floor, tax sleeves. You at least want to operate that way, even if you don’t yet.
  • You research before you buy. 10-Ks, on-chain data, earnings calls, primary sources — not just X threads.
  • You’ve been burned by your own behavior. Oversizing, FOMO, averaging down, holding losers. You know the loop. You want a system that stops you.
  • You refuse to pay 1% AUM. A private banker is $25K–$100K/yr to manage a portfolio you understand better than they do.
  • You hold across venues, asset classes, and currencies. Brokers, crypto, real estate, alternatives — in USD, EUR, GBP, JPY, MYR. The complexity is the point.

If most of these are you — Portivo isn’t for you yet.

  • You want someone else to drive. A robo-advisor or a real RIA will serve you better. Portivo assumes you’re the decision-maker.
  • You day-trade or scalp. Portivo is built around theses, reviews, and post-mortems on multi-week-to-multi-year horizons. Sub-week holding periods don’t fit.
  • You only own index funds. If your entire portfolio is VTI + bonds and you DCA, you don’t need a CIO. Vanguard will do.
  • You want to outsource thinking entirely. Portivo does the heavy reasoning, research, and remembering — but you still own the decision. If you want autopilot, this isn’t it.
  • You only want hot tickers, not trends. We help you spot signal early via ingest, scan, and watchlist — so you’re ahead of the cycle, not trading the headline. If “what should I buy today” is the only question, this isn’t the tool.
The problem

You don’t lack tools. You lack time and brainpower.

You already have the discipline. You know how to size, when to trim, how a thesis should read. The hard part is finding the hours — to read the transcript, triangulate it across three other things you saw, do the actual research, keep every thesis updated, and notice the trend before everyone is talking about it.

Three options exist today for someone managing serious capital. Each has a fatal gap.

Option 01The spreadsheet life

Run it all yourself in Notion, Notes, and a tab graveyard.

Reconciling positions, watching drift, drafting theses in Notes, tracking which account holds what. Three hours a week of busywork that compounds in noise — not in insight, and not in any Knowledge Base you can search later.

Gap → no compounding memory
Option 02The advisor option

Hire a private banker at 1% AUM.

$25K–$100K a year on a portfolio you understand better than they do. The relationship optimizes for their fee, not for the conviction you’ve actually built — and you still spend evenings on the work that matters.

Gap → outsourced judgment, kept work
Option 03Generic AI

Chat with ChatGPT or a portfolio assistant.

Helpful for a single question, useless across time. Chat tools answer in a vacuum: they don’t enforce sizing, don’t remember last quarter’s thesis, and don’t push back when you’re about to repeat the loss that hurt the most.

Gap → no policy, no continuity
How it works

Codify once. Compound forever.

Portivo is a small set of commands that compose into a workflow. Each one writes to the same Knowledge Base — so research feeds theses, theses generate reviews, reviews become post-mortems, post-mortems sharpen the policy.

STEP 01

Build the constitution.

A two-session interview — life, money, behavior, history — produces your Investment Policy Statement. The CIO derives the rules; you don’t fill in numbers. Sizing caps, cash floor, bucket structure, forbidden actions, behavioral profile.

STEP 02

Scan the world every morning.

Seven slots, enforced for diversity: a new macro theme, a sector you don’t cover, two company stories, a non-US story, two wildcards. The brief connects dots — supply chain to commodity to inflation to consumer — not just the headlines you’d see anyway.

STEP 03

Ingest anything you read or hear.

Drop a podcast transcript, a Substack post, an annual letter, a YouTube video, a Twitter thread, a research PDF. The CIO extracts the calls, agrees or pushes back with reasons, ties it to your existing positions and watchlist — then files it. Nothing dies in your tabs.

STEP 04

Research like an analyst.

Fifteen to twenty-five tool calls, six search passes, multi-tier evidence, red-teamed by adversarial personas. Every research memo answers the same checklist: what is it, who owns it, who’s selling, what breaks the thesis, what would change your mind.

STEP 05

Build the thesis. With math.

Secular or regime, with derived expiry. Hurdle rate vs. T-bills. Base / bull case grounded in milestones — not round numbers. ATR-based stops for regime trades. Fundamental invalidation for secular. Every tranche has a gate and a deadline.

STEP 06

Review on cadence. Versioned.

Theses get reviewed quarterly, monthly, or weekly — the CIO sets the cadence by type. INTACT / EVOLVED / BROKEN. New versions are saved; old ones are preserved. You can always read how your thinking actually changed.

STEP 07

Post-mortem every closed position.

Outcome × process. A profitable trade with broken process is still a failure. A losing trade with disciplined process is still good investing. Lessons feed back into your Investment Policy Statement — your behavioral profile sharpens with every cycle.

/build-ips · drafting your policyquestion 18 / 24
CIO
“Imagine checking your portfolio and it’s down 20% in a month. Be honest — what’s the first thing you’d want to do?”
YOU
“Add. I’d want to average down. I always do. I told myself last time I wouldn’t.”
DERIVED →
Behavioral pattern: averaging down on broken theses. New rule: tactical positions cannot be added to once invalidation is triggered. Forbidden action.
/scan · 7 slots filled09:14 ET
01 NEW MACRO · BoJ shifts JGB band — yen carry implications
02 SECTOR · US ag fertilizer crisis → soft commodity prices
03 COMPANY · NVDA earnings — guidance vs. hyperscaler capex
04 COMPANY · ASML new tool order book — non-portfolio
05 NON-US · Indonesia EV nickel export ban update
06 RADAR · Korean demographic cliff → consumption shift
07 RADAR · Energy storage breakthrough — too early to ticker
2 added to watchlist · 0 traded · 4 ingested to KB
/ingest · Invest Like the Best — Paul Tudor Jonestranscript · 14,200 words
CIO TAKE · 4 MAIN CALLS
1. US equities structurally bearish. Mkt cap / GDP 252%, S&P PE 22 → negative 10yr forward returns. Agree, calibrated.
2. Long JPY / Short USD/JPY. Yen “grossly undervalued 24mo” · Takaichi as Reagan analog · NIIP $4.5T mostly unhedged. High-conviction trade.
3. BTC > gold as inflation hedge. Two tail risks flagged: cyber warfare, quantum. Push back: BTC trades as high-β tech proxy, not digital gold.
4. AI as macro tail risk. “50–100M people die” before AI safety is fixed. Knightian — diversify, can’t position.
PORTFOLIO RELEVANCE →
PE-22 caution validates DCA discipline on your CSPX/IWDA deployment. JPY trade not IPS-compatible (file as conviction-builder). BTC framing reinforces gap-fill direction — 19.1% gold vs 0% BTC.
Filed: knowledge-base/analysts/invest-like-the-best/PTJ-2026-05.md3rd ILTB ingest
/research TAO — verdict22 sources · 5 passes
Decentralized AI compute is real, but the regime is hostile.
Subnet revenue $20M ARR and growing 40% QoQ. ETF pipeline 6–9mo. But BTC bear regime + alt liquidity drying = expect drawdowns regardless of fundamentals.
Bull case · 5 pillars5/5 documented
Bear case · 5 pillars5/5 documented
Source credibility avg74 / 100
VerdictBULLISH · moderate
Verdict-flipper: subnet emissions exceed organic demand for 2 quarters.
/build-thesis TAO · v1SECULAR + catalyst
Decentralized AI compute, held through cycle weakness.
Hurdle vs T-bill (5.1%)+22% needed
Base case+85% over 18mo
T1 milestone$50M ARR · ~$X
Catastrophic floor−65% from entry
Tranches3 · gates + dates
Pre-mortem: most likely failure = ETF doesn’t materialize AND BTC stays bear → exit at T1 anyway.
/update-thesis TAO · v3EVOLVED · review 06.18
Elementv2 → v3?
Viewintact
Catalyst (ETF)delayed Q3→Q4!
T1 milestone$50M → $42M ARR~
Capital justifiedweakening!
VerdictEVOLVED
v3 lowers T1, extends expiry by one quarter, and trims position 25% — opportunity cost argument is weakening with no milestone progress.
/post-mortem SOLCLOSED
OUTCOME
Good · +41%
PROCESS
Sloppy
“Profitable trade. Broken process. Still a failure to learn from.”
Tranche 3 added without gate met. T2 hit and not trimmed. Pattern from 2024 repeating: can’t sell winners. IPS amendment proposed: trim discipline rule for tactical winners >2x entry.
Step 01 / 07· click any step to jump
Wired into the world

The CIO calls real data — not vibes. Every claim is sourced and verifiable.

PricesYahoo Financelive quotes · profile · 52-week range
TechnicalsTwelve DataATR · EMA · RSI · MACD · weekly + daily
FundamentalsFinancial Modeling Prepincome · balance · ratios · estimates
Macro · USFREDrates · CPI · GDP · yield curve
Macro · GlobalDBnomicsIMF · ECB · BIS · OECD · World Bank
DeFiDefiLlamaTVL · protocol revenue · stablecoin flows
CryptoCoinGeckomarket data · token resolution
PoliticsCapitol Tradescongressional disclosures · STOCK Act
WebLive web searchmulti-pass · cited · tier-ranked
Your libraryKnowledge Basesemantic + structured search of your filings
A Knowledge Base that thinks with you

The CIO already knows what you’ve read.

You don’t ask Portivo to “synthesize three articles.” You do your normal work — research a name, update a thesis, ask a question — and the CIO connects the dots from your Knowledge Base on its own.

01

Indexed two ways.

Every transcript, memo, voice note, article, thesis, post-mortem — searchable by meaning (vector) and by structure (ticker, theme, date, source). Same content, two retrieval paths.

02

Pulled in automatically.

Run /research USD-JPY or just ask “start me on USD weakness.” The CIO checks your library first — before going to live data — and brings forward what you’ve already saved.

03

Connected, with citations.

Druckenmiller said this in April. Marks framed it in March. Your own voice memo pre-empted both. You never told it to triangulate — it just does, because that’s what a CIO does when they know the book.

YOU“Start me on USD weakness — where do I begin?”Apr 14 · sparring
CIO checks your Knowledge Base first — semantic + structured — then the data
WHAT YOU ALREADY KNOW · 4 RELEVANT THREADSretrieved from your KB
From your KBDruckenmiller · ILTB Apr 2026 — fiscal-driven USD weakness, prefers JPY on NIIP + valuation. You ingested this 5 days ago.
From your KBHoward Marks · The Long View — same regime call, prefers quality non-US equity. Different horizon, same direction.
Your own note“JPY — something here?” · voice memo Apr 09. You caught the BoJ band shift before the desks did.
Your thesis historyNo active FX thesis. Last USD short was 2022 — closed early, post-mortem flagged “exited the regime trade too soon.”
CIO takeThree smart sources agreeing, your own note pre-empting them, FX bucket open in your IPS, and a behavioral lesson from 2022 worth remembering. Want me to run /research USD-JPY against your book?
Beyond the workflow

Not an assistant. A sparring partner.

The structured commands are the spine. But you can bring the CIO anything — a half-formed thought, a thesis you want stress-tested, an urge you can’t shake. It will challenge you with your own rules, your own history, and the data.

CHALLENGE“I want to add to NVDA. Talk me out of it.”
PRE-MORTEM“What’s the most likely way my TAO thesis fails?”
STRESS TEST“Run my book through a 2008 scenario.”
DEVIL’S ADVOCATE“Argue the bear case on my biggest position.”
BEHAVIORAL“Am I about to repeat 2024?”
REGIME“Is BTC accumulation or distribution right now?”
DECISION“Should I trim ASML at T2 or let it run?”
RECALL“What did Druckenmiller say about Japan last month?”
OPEN“I’m worried about the dollar. Where do I start?”
portivo · sparring11:42 ET · Tue Apr 14
YOU
“I want to add to SOL. It’s down 18% from last week — feels like a gift.”
CIO
Before we discuss adding, three things from your own record.
IPS · §1 SIZINGMax single position 8%. SOL is currently 9.4% — already in violation. Adding would extend the breach.
THESIS · SOL v3 · ACTIVETactical regime trade. T1 hit Mar 22 — the IPS-required trim was not executed. The position is over-sized vs your own gates.
POST-MORTEM · 2024 SOL CYCLEYou filed this exact pattern last cycle — averaging down on a tactical position after a drawdown. Result: held through invalidation, exit was forced. Behavioral tag: “can’t sell winners → averages losers.”
The thesis hasn’t changed. Your sizing has. The honest action here is trim back to 8%, not add. If you want to argue the other side, tell me what’s different from April 2024 — I’ll listen.
YOU
“Fine. But what would have to be true for adding to make sense?”
CIO
A new thesis. Specifically: SOL revenue or DEX share crossing a level you defined in advance — not a price level. Then it’s a fresh entry under a fresh policy bucket, not an add to a tactical position you should be trimming. Want me to draft that thesis, or close this loop?
Inside the product

This is what your book looks like.

Portivo is a real app — not a chat window pretending to be one. Every position, every thesis, every closed trade lives as a structured document with its full audit trail.

Investment Policy Statementv.2026.04

1. Sizing & concentration
Max single position8% of investable
Half-size on entry30-day review
Asset class cap (crypto)15%
Tactical positionsmax 5 · ≤25%
2. Cash floor
Floor target$80,000
Below floorbuy-freeze
3. Thesis discipline
Mandatory fieldsview · catalyst · invalidation
Hurdle vs alternativerequired
Review cadenceCIO-set by type
4. Forbidden actions
Average down on broken thesisblocked
Add to tactical past invalidationblocked
Trade without thesisblocked
Your policy is the law

Your Investment Policy Statement. Enforced on every recommendation Portivo makes.

Most AI tools treat your portfolio as data. Portivo treats your policy as law. Every suggestion is checked against the document on the left — and shipped with the receipts.

Edit it as a doc. Fork a template. Grow it month by month from your own post-mortems. The agent reads the same document, the same way, every morning.

No single position > 8%Sized at 7.2% post-trade7.2 / 8.0
Cash floor maintained$84.2K vs $80K target+$4.2K
Tactical count4 of 5 · combined exposure 21%4 / 5
!Hurdle margin tight+6% over T-bill — borderline justified+6%
Thesis review current3 of 4 pillars hold; reviewed 2026-04-29v3
Your year with Portivo

A real cadence. Not a chat habit.

Most “AI for investing” tools assume you’ll remember to use them. Portivo runs on rhythm — daily, weekly, monthly, quarterly, and the things that only happen when they happen. Each cadence has its own work and its own document trail.

Daily

Every morning

~10 min · coffee in hand
  • /scan7-slot brief. Macro, sector, companies, non-US, wildcards — connected to your book.
  • driftPosition drift vs your sizing caps. Quiet alert when something’s creeping.
  • watchlistRe-flag boost for items the CIO has surfaced before.
Weekly

Every Sunday

~30 min · the wind-down
  • portfolioNAV, IRR, PnL vs your custom benchmark — not the S&P by default.
  • /researchDeep dive on one ticker the watchlist promoted this week.
  • budgetCashflow + cash floor check. Savings rate vs target.
Monthly

First Saturday

~60 min · the audit
  • violationsPolicy review. Severity, recurrence, behavioral patterns — surfaced.
  • behaviorBehavioral tags on every close. Loops surface before they’re habits.
  • /update-thesisRegime theses reviewed at their cadence. Versioned.
Quarterly

Every quarter-end

~half day · the review
  • /update-thesisAll secular theses. INTACT / EVOLVED / BROKEN.
  • policyIPS amendments from this quarter’s post-mortems folded in.
  • CIO noteQuarterly letter — what worked, what didn’t, why.
On event

When it happens

as needed · minutes per
  • /ingestWhenever you read or hear something good. Transcript, article, voice memo — filed and connected.
  • /build-thesisNew position. With math. Policy-checked before you size.
  • /post-mortemEvery full close. Outcome × process. Lessons amend the policy.
“The discipline of an institutional CIO, applied to my portfolio, at the cost of a software subscription.”
— Portivo, in one sentence
Frequently asked

The questions sophisticated investors actually ask.

Who is Portivo for, exactly?+
Self-directed investors managing $250K–$10M who already think in policy: position sizing, exposure caps, written theses, post-mortems on closed positions. Tech founders, operators, finance professionals who run their own book. If you’re managing material capital and want institutional discipline without a 1% AUM fee, Portivo is built for you. If your portfolio is index funds and you DCA, you don’t need this.
Is Portivo an advisor? Will it pick stocks for me?+
Portivo is software for managing your own portfolio under your own policy. It surfaces what matters, researches what you ask, and checks every action against your IPS — but you make every decision. We do not custody assets, we do not provide personalized investment advice, and we do not push trades.
How is this different from other AI portfolio tools?+
Most are conversational portfolio assistants — chat with an AI about your money. Portivo is a workflow harness. Every recommendation is checked against a written policy you authored; every research memo is structured, sourced, and red-teamed; every thesis carries a hurdle rate and an invalidation; every closed position generates a post-mortem that amends your IPS. The product is the discipline and the compounding knowledge base, not the chat.
When can I use it?+
Private beta is opening to a small group of self-directed investors over the coming weeks. The first few seats include a guided onboarding where you write your IPS together with the CIO — it interviews you, drafts the document, and you edit until it’s yours. Join the waitlist and we’ll be in touch.
How is Portivo regulated? What does it not do?+
Portivo is software you operate yourself. It is not a registered investment adviser, broker-dealer, or financial planner; it does not custody assets, route orders, or provide personalized investment advice. Nothing it generates — scans, research, theses, post-mortems, sparring transcripts — constitutes a recommendation to buy, sell, or hold any security. All output is informational, based on data sources you can verify, and intended to support your own decision-making. You are solely responsible for every action you take, and you should consult a licensed adviser, accountant, or tax professional before acting on anything you see here.
Important

Portivo is software, not a financial adviser. It is a tool that self-directed investors use to organize their own research, write their own policy, and review their own decisions. It is not a registered investment adviser, broker-dealer, or financial planner.

Portivo does not custody your assets, place orders, or provide personalized investment advice. Nothing produced by the product — scans, research memos, theses, post-mortems, sparring transcripts — is a recommendation to buy, sell, or hold any security or asset. All output is informational only, generated from public data sources, and intended to support your own decision-making.

You are solely responsible for every investment decision you make. Consult a licensed adviser, accountant, or tax professional before acting on anything you read here. Past performance is not indicative of future results.

Private beta · 2026

Built for the way you already invest.

Join a small group of self-directed investors helping shape Portivo’s first public release.