The executive cognitive operating system.

Not a productivity app. A cognitive layer that compounds a model of how you think — and gives the bandwidth back to the decisions only you can make.

02 · the drop

Drop anything.
Cited verbatim.

Your ChatGPT exports. Voice memos about who you are. The note you wrote when last year was hard. A 4-minute brief telling the AI about you. Already paid for ChatGPT? Already spent six months training Claude on you? Already have an Obsidian vault, a Notion workspace, ten years of email? Drop it in. Timed reads what you already taught the others — then builds a model exclusively for you. You don't need 12 months to feel like it knows you. Supercharge it on Day 1 with everything you already have.

txt
ChatGPT export
18 months · conversations
txt
Claude export
projects · threads
m4a
Voice memo — about you
intro brief · 4m 12s
m4a
Voice memo — board prep
06:48 · 2m 41s
md
Notes on yourself
anything you've written
txt
Year-end review
what last year was about
pdf
Board pack — Q2
47 pages · upload
eml
Counter-offer thread
14 msgs
→ Timed
+ 3.2K obs · Tier 0 seeded
surfaced · 09:14 · tuesday

"The Mott counter is the call your week is about — they soften every fourth response. You've avoided this thread for nine days."

source: mott-counter-thread.eml  ·  msg 14 / 14  ·  ¶ 3  ·  embedded 2026-05-17 02:14 UTC  ·  Tier 0 → Tier 1 (daily) → Tier 2 (signature: "counter-party softens before concession")

Every line cites the exact paragraph it came from. Not AI-extracted. Quoted. Verbatim. By construction.

0·0
35,247 micro-decisions a day · 9 that move the company · the rest is tax
03 · you don't have a productivity problem

Fewer decisions. Better ones.

Timed pre-decides everything that's pre-decidable from your context — so the 9 land with the bandwidth they deserve. The other 35,238 stay handled, in the background, by the system that has slept on your inputs.

the diagnosis · not productivity

You don't have a productivity problem. You have a cognitive bandwidth problem. Timed builds a model of how you think.

The premise — that drove every decision in the system Architecture: Stanford Generative Agents · Park et al. 2023
a tuesday morning

Surfaces the one call your day is really about.

The morning briefing is not a feed. It is one decision.

The system reads your week, your inbox, your governance commitments and your family calendar — runs nightly reflection — then, by 09:14, it knows which call your Tuesday is really about.

It will not draft a reply. It will not move a meeting. It will not contact anyone. It reads, reflects, recommends. You decide.

  • The Mott decision — surface it now.Highest signal
  • ~14:00 internal review — push to Thursday.Defer
  • 17:30 concert — protected.Protected

Generated by the strongest reasoning model on Earth from 14,247 indexed observations the night before. By 09:14, one sentence.

Timed — morning briefing — 09:14 AM
09:14Mott decision — the call your week is about. Their counter softens every fourth response.
10:30Board prep block — calendar holds.
14:00Internal review — defer to Thursday.
17:30Lila's concert — protected.
22:00Outcome capture — log the day.
voice orb — 06:48
orb"You've avoided this for two weeks. Want me to walk you through what you keep flinching from?"
06 · the voice orb

Talk to it. Like a chief of staff.

No commands. No syntax. A single orb listens, reflects, offers back what it heard.

The capture surface is conversation, because thinking out loud is how executives already work. Off-record turns leave no trace. On-record turns feed the nightly engine.

orb · this morning

"You've avoided this decision for two weeks. I have the data. Want me to walk you through what you keep flinching from?"

"You've avoided this Mott decision for two weeks. I have the data. Want me to walk you through what you keep flinching from?"
22:30
while you sleep

Tonight, while you sleep.

The nightly engine runs on biology — four sleep stages, four parallel actions. By breakfast: one sentence.

23:00 NREM 1–2 brain catalogs the day Timed scores importance
01:30 NREM 3 · deep hippocampus replays at 20× Timed extracts patterns
03:30 REM cortex weaves themes Timed synthesises traits
05:00 outcome predictions calibrate Timed updates rules

"Move the Q2 board prep to Tuesday morning. Don't move it again."

why now

The class of model that can actually think arrived last year.

2024 Recursive nightly reflection at executive depth was a research artifact.
2026 A single user's full night of synthesis now costs less than a coffee.
18 mo The window where the moat is built — or never is.
month-over-month fidelity

Month 6 is sharper than month 1. Not as marketing — as math.

A competitor launching tomorrow starts at Month 1 against your Month 12. They can't catch up by ingesting more data faster, because the data is exclusively yours. The system that has slept on your inputs for a year is the moat.

M1 M3 M6 M12

Tier 3 personality traits update monthly with precision scalars (0.5–1.0 confidence) and bi-temporal valence vectors. A competitor can't backfill this — your data is exclusively yours.

10 · the doctrine

It will never act for you.

Observes.
·
Reflects.
·
Recommends.

Never sends mail. Never books. Never replies pretending to be you. The most dangerous AI right now is the one that does those things. Timed will never be that. Audited at every prompt.

And once it surfaces a recommendation, it cannot silently retract it. If a recommendation changes, you see why. Never-retract · enforced by the 10-check adversarial CCR audit.

doctrine.log · live
REQUEST: send email on your behalf · DENIED REQUEST: book meeting · DENIED REQUEST: reply to thread · DENIED REQUEST: contact stakeholder · DENIED REQUEST: move calendar event · DENIED
Adept raised $415M to build agents that act on the world. Acqui-hired by Amazon, 2024.
The market is not ready. The never-act doctrine is the credible alternative.
11 · built like infrastructure

This is real software. Built to the standards the operator deserves.

GDPR export · cyber guarantee · your data never leaves the trusted boundary.

99.97% uptime · 30 d
STATUS · OK
0Supabase Edge Functions in productionDeno · service-role hardened · per-route RLS
0swift-testing assertionsgreen on every push to unified
0recursive reflection tiersobs → daily → behavioural → personality → rules
0embedding models, hot + coldVoyage voyage-3 1024-d · OpenAI 3-large 3072-d
0calendar providers, one schemaApple EventKit · MS Graph · Google Calendar
0OAuth providers wiredApple · Microsoft · Google · magic-link
0guards before any LLM callmodel allowlist · token cap · prompt fencing · scoped RLS
0particles in the voice orb @ 60 fpsCanvas2D · DPR-aware · IntersectionObserver-gated
0tokens of you, hot-loaded into Opus 4.7prompt-cached · sub-second cold-start
0nightly Opus 4.7 reflectionTrigger.dev v4 · scheduled 03:00 local · idempotent
0third-party API keys on the deviceevery key proxied · ephemeral WS tokens · 60 s TTL
Edge function deployed · voice-llm-proxy 2s ago
14 · traction

The day-one users are the founders.

Yasser — billion-dollar operator across property, OTR, motorsport — uses Timed as his actual executive cognitive layer. Right now. Every morning.

Ammar — the engineer — is the second user. Every night the system reflects on both. Every morning the briefing is the product.

The dogfood loop is the entire team. Every shipped feature has to survive Yasser's Tuesday.

06:48
Yasser opens the voice orb — pre-coffee reflection.

Built by father & son. A billion-dollar retail operator and the engineer building inside his head. Mac & iOS.

Join the early access

The first cognitive OS ever. Made in Adelaide.

Timed Yasser Shahin · Ammar Shahin © 2026