Good evening, there.
Your AI learns how you work β transparently, privately, and under your control.
Recent chats
Chats & Projects
Resume work with the correct project scope, profile overlay, and learning permissions.
DCS Engineering OS
Control Tower, Adaptive Workspace, Connector OS, Mission OS, and R-Series integration.
AIE Workspace
Final UI design contract, profile controls, memory, intent, receipts, and developer handoff.
TRD Mind & Mart
Private AI workspace and open AI marketplace connected through TRD Console.
Recent conversations
| Conversation | Project | Scope | Learning | Updated |
|---|---|---|---|---|
| No conversations yet β start a chat and it will appear here. | ||||
My AI Profile
Everything the system has learned is reviewable, editable, exportable, and receipted.
Applied
Shapes every answer Β· β₯ 0.90Suggested
AIE thinks this Β· NOT acting on it Β· 0.70β0.89AIE inferred these but has not confirmed them with you β so it shows them and refuses to use them until you say yes. Every other memory silently acts on its guesses; this one does not.
Communication
AppliedWork Intelligence
Read overlayExecution Rules
User Controls
Learning Timeline
Point-in-time replay over the existing Time Machine receipt and version chain.
Events
DARK fixtureUPDATE Β· Quality before speed
Explicit confirmation raised confidence to 0.99 Β· receipt aie_rcp_9f2a
LEARN Β· Final documents in English
Cross-session recurrence Β· confidence 0.96 Β· global scope
SUPERSEDE Β· Planning language
English-only β Hindi-English planning Β· old validity window closed
DECIDE Β· Intent threshold
Consequential-action abstention remains 0.85 Β· config change receipted
DELETE Β· Temporary launch date
Tombstone created Β· history retained for proof
Point-in-time profile
11 Jul 2026 Β· 18:24Memory & Knowledge
Bi-temporal facts, episode summaries, entities, and prevention rules. Raw conversation turns are never stored.
Active facts
Receipted writesCorrections & Prevention
Mistake β correction β reason β prevention rule β measurable accuracy improvement.
Active prevention rules
βLoading your prevention rulesβ¦
Feedback fusion
Confidence v2Correction history
Mistake: generated an image after the user was discussing prompt quality. Correction: classify quoted examples as discussion unless explicit execution is present.
IF quoted_example=true AND explicit_execute=false THEN intent=discussion; action=abstain_from_executionCorrection: use the approved TRD asset automatically for every TRD project surface.
scope=project:trd Β· asset=trd_logo_locked Β· replacement=blockedIntent Lab
Test discussion vs instruction, quoted examples, abstention, and correction-derived prevention rules.
Classifier
No execution. Quoted/example span excluded by Boundary B5.
Execution thresholds
Config Β· DECIDE receiptsAdversarial restraint suite
Merge-blockingTransparency
See what was learned, why it was learned, the evidence used, and every change in its validity window.
Why this preference?
βLoading your highest-confidence beliefβ¦
Confidence explanation
βClick a confidence badge on any fact to see why β occurrences, confirmations, contradictions, recency, and the receipts behind it. Every value is fetched from /aie/explain; nothing here is hardcoded.
Learning boundaries
Zero-leak gateSession facts expire. Durable promotion requires recurrence or explicit pin.
Regex + model redaction before persistence; content never logged.
Single occurrence remains evidence-only.
Learn pipeline aborts at step 0.
Quoted, pasted, and example spans are non-executable.
Assistant turns never become user beliefs or opinions.
Receipt Explorer
Every LEARN, UPDATE, SUPERSEDE, DELETE, and DECIDE mutation carries a receipt ID.
Receipt list
Demo receipts β unsigned; fleet verification at bindingPrivacy & Consent
Consent is enforced at the pipeline mouthβnot only in the UI.
Learning controls
Structural guarantees
Category exclusions
Connectors & Warm Start
Warm-start imports are consent-gated, boundary-filtered, and receipted.
Import rules
P5 Β· DARKRecommendations
High-confidence suggestions only. Every dismissal is respected and anti-nag cooldown is enforced.
Keep Preview collapsed until an artifact opens
You usually work in a wide chat surface and open the right panel only for documents, HTML, code, or images.
Connect GitHub for project warm start
Import repository names and frameworks after consent. Engineering Memory remains the source of truth.
Use High quality for architecture and UI handoffs
Your corrections show that completeness and consistency matter more than fast first drafts for complex work.
Render-test before delivering frontend artifacts
Automatically run desktop-state checks for collapsed sidebar, hover expansion, preview resizing, and settings.
Recommendation policy
Anti-nag activeHealth & Benchmarks
Current internal build status. All flags remain DARK; DK is the only flag flipper.
Phase status
Binding queue
8 binding itemsEvaluation gates
How AIE Works
AIE learns what you prefer, shows you everything it believes, and can prove where each belief came from. This page explains exactly how β including what it will not do.
The one-line version
You talk to AIE normally. When you state a standing rule about yourself β "always reply in English", "never use emojis", "confirm before deploying" β it remembers. Every other product asks you to fill in a settings page. AIE learns it from the conversation, then shows you the receipt.
It does not remember your questions, your one-off requests, or anything you did not say about yourself.
What it will remember
Durableprefers_language.project_stack, never as "you like Rust".What it will refuse to remember
Blocked in codeConfidence β and why AIE stays quiet when it is unsure
The productEvery belief carries a confidence score. What AIE does with a belief depends entirely on that number:
This gap is the whole point. A memory system that acts on a 0.72 guess is worse than one with no memory at all β it is confidently wrong about you, and you cannot see why.
It forgets, on purpose
A belief AIE only inferred loses confidence over time β half-life 90 days. If you stop behaving that way, it fades.
But a preference you stated out loud never decays. Silence is not retraction. If you told AIE something once, it does not quietly forget because you did not repeat yourself.
And decay has a floor (0.30) β it can weaken a belief, never erase the evidence that you said it.
Correcting it
When AIE gets something wrong, tell it β in Corrections & Prevention. A correction can become a prevention rule: a standing instruction the intent gate checks before doing anything consequential.
Rules are matched by meaning, not keywords. "Never replace locked assets" also fires on "swap out the pinned files".
If the matcher cannot decide, it blocks. A safety rule that fails open is not a safety rule.
You can check everything
ReceiptsWhat is NOT built yet
Honest listBecause a manual that oversells is the same lie as a button that does nothing:
Every tab, and what it does
Audited β every control clickedaie_sessions has no column for them, and a filter that always returned nothing would be a lie about your data.Try it in sixty seconds
always reply in Englishprefers_language: English is now there, with a confidence and a receipt.