Ultra-compact multi-expert AI for systems that must keep learning.

Add new domain experts at near-zero parameter cost. Permanently protect everything the model has already learned.

≈10²
params / expert
0
prior weights touched
n+1
experts, same core

Arch/frozen core + additive experts

Target/MCU / edge NPU

Regime/sequential · class-incremental

01 / 05

Problem

Learning is easy. Not forgetting is hard.

Edge and industrial systems accumulate tasks over months and years. Each update to a monolithic model puts everything it already knows at risk.

Cost
Full retraining is compute- and data-hungry, every single time.
Regression
New tasks overwrite old behavior — catastrophic forgetting.
Assurance
Safety-critical systems need structural guarantees, not empirical hope.

Fig. 01 — retention after learning task B

Monolithic retrain

Task A degraded — weights overwritten.

VCTR04AI multi-expert

Task A and Task B experts coexist. Both intact.

02 / 05

Solution

Knowledge kept physically apart.

Instead of rewriting shared weights, VCTR04AI grows sideways: each capability lives in its own frozen expert, and the core model stays untouched.

  1. 01

    Observe

    A lightweight router reads the incoming distribution and identifies which experts are relevant.

  2. 02

    Add

    A compact expert — a few hundred parameters — is trained for the new task alone.

  3. 03

    Lock

    Existing experts are frozen. They are never merged, averaged, or re-optimized.

  4. 04

    Route

    At runtime inputs dispatch to the right expert blend, preserving accuracy across every task learned so far.

Fig. 02 — expert stack over time

Core
E1
E2
E3
E4
n+1

Core weights frozen

Δparams / task ≈ 10²

03 / 05

Key properties

Built for systems that cannot afford to forget.

P-01

Extreme parameter efficiency

New experts cost a few hundred parameters, keeping the architecture viable on microcontrollers and small edge accelerators.

P-02

Mechanical isolation

Prior experts are frozen and never updated. Forgetting is structurally prevented, not merely discouraged by a loss term.

P-03

Memory-constrained design

Built from the ground up for devices with limited RAM, flash, and power budget — not scaled down from a datacenter model.

P-04

Safety-sensitive by default

Isolation, traceability, and deterministic routing make the system easier to validate, audit, and re-certify.

04 / 05

Current status

Prototype stage, stated plainly.

No benchmarks-by-press-release. Here is exactly what runs today and what is still under construction.

  • Multi-expert prototypesWorking
  • Isolation on sequential benchmarksStrong
  • Parameter-efficient expert additionWorking
  • Routing at known task boundariesBasic
  • Automatic routingIn development
  • Production runtimeIn development
  • Edge deployment toolingPlanned

05 / 05

Contact

Let's talk.

Research partnerships, evaluation pilots, licensing questions. Every message is read by a human.

samat@vctr04ai.com

One email at a time · unsubscribe any time