Why it compounds
Context is the moat.
Every operation can host the same open weights models. Context is what compounds.
01 CAPTURE
Engram
Dedicated inference for your sovereign AI models.
Train sovereign AI models in their weights and in their memory, on premises and in air gapped environments.
Hyper Engram turns the work an operation does into training signal for the open weights models it hosts. Runs, outcomes and corrections are captured as episodes, compressed into versioned principles and applied two ways: post training into the model's weights, and agentic memory the model reads at inference. Training data, weights and memory never leave the perimeter.
Recursive improvement, one loop at a time.
A master agent runs each RSI Loop. Two routines distil and verify what was learned every day. They never call each other directly: each reads and writes one shared memory, which the model reasons over at inference.
Four stages. One training platform.
Capture, compress, train and reinflate. Every stage runs inside your perimeter, on premises or air gapped, on models you host.
Capture
Every run becomes training data. Each run is kept as an episode with its outcome, its feedback and the corrections a reviewer made, inside your perimeter.
Compress
Distillation into principles. Daily routines distil raw learnings into versioned principles that cite their sources. Contradictions are resolved and what never mattered fades.
Train
Into the weights. Curated episodes and principles become post training data for the open weights models you host, trained on your own hardware.
Reinflate
Into memory, at inference. The model reads the compressed memory and expands it into the problem in front of it: this site, these regulations, these people.
Learned tonight. Used tomorrow morning.
Each routine opens its run by reading the feedback and gap records on its newest episodes. A rule recorded today changes tomorrow's behaviour, with no code change.
A run misses something
A design senses the warning a real incident would have started with, and never says who must act on it.
A later reader finds it
The master agent reads the run and files a gap record on its episode: what was missed, the fix, and a rule for next time, written in the routine's own terms.
The routine reads before it acts
Its intake lists every rule in force. The rule is now part of how it works, with no change to its code.
The record closes, or it does not
The next RSI Loop checks whether the gap closed. A record still open is the first item of the next plan.
// filed on the routine's newest episode { gap: "designs sense the warning but never name who acts", fix: "brief a routing principle from general practice", found_by: "master agent", status: "open", rule_for_next_time: "every sensed warning names the role that acts, the reading that starts the clock and the options of the decision" }
Every learner in the system, from the routines to the fleet, keeps its episodes in this one shape, so any reader can correct any writer.
Compress once. Reinflate anywhere.
Eight raw learnings are compressed into one principle, which the model reinflates into designs for operations it has never seen.
- Benchmark missA conveyor design sensed belt misalignment and never said who acts on it.
- Engineer correctionThe alarm went to the control room, which cannot stop a haul truck. Route it to the pit supervisor.
- Real requirementA tender asks for alerts to reach “the responsible officer” within a minute. Responsible for what?
- Field learningOperators ignore an alert that arrives without a decision attached to it.
- Benchmark missA gas reading crossed its limit. The design logged it and paged no one.
- Engineer correctionName the reading that starts the clock, not only the threshold.
- Earlier noteSend every alert to the shift manager.
- Earlier noteMake critical alerts red.
Press Compress. Six learnings become one principle, one contradiction is resolved and one note fades because it never mattered.
- Waiting
- Compress, then pick a place. The same principle becomes a different design in each one, because the world it lands in is different.
Illustrative. The principle is one the platform learned this way; the learnings and places are drawn for this page.
Two training targets. One platform.
Engram places each lesson where it does the most good. Stable skills used at volume are trained into the weights. Knowledge that changes with the operation is kept in memory, where it stays current, readable and reversible.
| Into the weights | Into memory | |
|---|---|---|
| Best for | Stable skills used at high volume | Knowledge that changes every month |
| A new regulation arrives | Joins the next training run | One new entry, verified and live within hours |
| Reading what was learned | Measured through evaluation | Every principle lists the learnings it came from |
| A stronger base model arrives | The training run is repeated on it | The memory carries over on day one |
| Undoing a lesson | Roll back to an earlier checkpoint | Retire one versioned principle |
| Where it runs | Inside your perimeter, on models you host | Inside your perimeter, on your infrastructure |
Measured every loop.
Praxis, trained on Engram, is measured on PADI, our benchmark for physical AI system design, against frontier AI on its own, on fresh tasks every RSI Loop.
Praxis, RSI Loop 22, blind
Frontier AI on its own, same tasks
Unseen tasks where Praxis came out ahead
Industries measured, in three countries
Checks passed as a share of checks judged, each RSI Loop. Loop 22 is a blind evaluation on tasks the platform had never seen; earlier loops reused some tasks and show the trend. Results are public; the method stays private. Every figure on PADI →
Eight rules every loop keeps.
They hold across every phase and every agent. A change that would bend one stops for human review.
Train where it belongs
Stable skills go into the weights. Knowledge that changes goes into memory.
Meaning, never keywords
Any decision about what a sentence means is put to a model and stored. Keyword scoring is banned.
Every claim stands on a source
An entry needs a dated source. A principle needs the learnings it was distilled from.
Evaluation stays blind
Held out tasks are never shown to the planner. Incident cases never feed the memory.
Two agents, two views
Every writer has an adversarial verifier. Nothing verifies its own output.
The record is the context
Commits, episodes and gap records carry the why. A session that lost its memory reads the log first.
Every release is reversible
Each RSI Loop is tagged. A bad change is reverted by its own commit, a bad principle by its own version.
Sovereign by design
Training data, weights and memory stay in your perimeter, on premises or air gapped.
Read how it works.
The argument, the architecture, the measurements and the product, each on its own page.
Compression and Reinflation
Compression, reinflation, and why what a system learns should survive the next model.
Muhammad Umar · 10 Oct 2026 →ArchitectureAgentic Memory Architecture
Memory told as a bow tie: experience, a waist of nightly distillation, and context assembled for each design.
23 Oct 2026 →BenchmarkPhysical AI Design Index
Fresh tasks across 13 industries in three countries, Praxis against frontier AI on its own.
Updated every RSI Loop →ProductMemory at work in Hyper Pragma
The same memory inside a product: every run graded, every lesson visible, every lesson revocable.
Hyper stack →What is Hyper Engram?
Hyper Engram is CodeNinja's platform for training sovereign AI models on premises and in air gapped environments. It trains open weights models in their weights and in their memory, from the runs, outcomes and corrections of the operation they serve.
How does Hyper Engram train a model?
Runs are captured as episodes and compressed daily into versioned principles that cite their sources. Curated episodes and principles become post training data for the model's weights, and the same principles form an agentic memory the model reads at inference.
What is compression and reinflation?
Compression turns many observations into one small principle, resolving contradictions and dropping what never mattered. Reinflation is the model expanding that principle into a new situation: a different country, operation or set of people. The idea follows Michael Levin's work on self improvising memory in living systems.
What goes into the weights and what into memory?
Stable skills used at high volume are trained into the weights, where they are fastest and cheapest to serve. Knowledge that changes with the operation, such as regulations, equipment and procedures, is kept in memory, where it is current within hours and can be read and reversed.
What happens when a better base model is released?
It is swapped in. The memory carries over unchanged, and the curated training data is applied to the new base model.
Where does Hyper Engram run?
Inside the customer's perimeter: on premises, in a private cloud or fully air gapped, on models the customer hosts. Training data, weights and memory never leave it.