Skip to main content
SkyDynamics - homeSkyDynamics

AeroBrain by SkyDynamics

AI you can audit. Decision support from approved manuals and live data, with every answer traced to its source.

AeroBrain is the SkyDynamics set of high-assurance AI technologies for aerospace, defence and other safety-critical fields. It is built to read real-time avionics data against approved manuals and to propose options that a person checks and decides on.

Request a technical briefingSee the technologies
Animated diagram: the live avionics data of an aircraft shows the vibration of engine 2 rising. LLiDA detects the trend, the Contextual Composition Protocol understands the situation, the RAG engine retrieves the evidence and the Assurance Small Language Model proposes three possible solutions, inside a fixed time budget. The crew decides.

Aircraft avionicsLive

Illustrative

Engine 2 vibration

Rising

The crew decides. Everything is recorded.

  1. LLiDA

    Detects

    Trend found

  2. Protocol

    Understands

    Climb, icing

  3. RAG

    Retrieves

    4 sources

  4. SLM

    Proposes

    3 options

Time budgetAnswered inside the budget

Possible solutions

  1. Apply the high engine vibration procedureSource: Operator QRH
  2. Reduce thrust on engine 2 within limits and monitorSource: FCOM limits, live data
  3. Review diversion options if the trend continuesSource: Operations manual

Audit record written: data, sources, model version, decision

High assurance
Developed to MISRA C and the objectives of DO-178C, with requirements, code, tests and evidence traceable in both directions.
Auditable
Every input, source, model version, output and decision is recorded, so any answer can be replayed and examined afterwards.
Real time
Designed to work on live data from the avionics and to answer while the event is still unfolding, not after the flight.
Hard real time
For functions where a late answer counts as a failure, each stage is designed to run inside a fixed time budget. The figures depend on the hardware and are set with the customer.

  • Topproduct

  • Mostinnovative

  • Innovationtech

  • Disruptivetech

  • Highassurance

Use case: real-time decision support on the aircraft

The same event, stage by stage. Four technologies working together, and a crew that decides.

Engine vibration rises in the climb: from live avionics data to possible solutionsIllustrative

The event

In the climb, in icing conditions, the vibration of one engine starts to rise.

Advisory thresholdNormal band
Phase
Climb
Signal
Engine 2 vibration
Source
Avionics, live
Conditions
Icing

  1. LLiDA captures the trend

    Engine parameters arrive from the avionics in real time and are sampled into memory at a fixed rate, on the aircraft. The rise is seen as it happens, not after the flight.

    Vibration above the normal band and rising

  2. Contextual Composition Protocol holds the situation

    Phase of flight and aircraft configuration from the avionics, open MEL items and weather are already in memory. The event is read against them.

    Climb, icing conditions, no open engine items

  3. Secured RAG engine retrieves the evidence

    The live readings and the passages of the operator's approved manuals that apply to this situation are retrieved deterministically, each with its source and time.

    Live readings, procedure, limits and MEL reference

  4. Assurance Small Language Model proposes the solutions

    A compact, auditable model composes the possible solutions from that evidence only. Each one keeps its source.

    Three possible solutions, each with its source

  5. Governed layer the crew decides

    The solutions are offered, never executed. The decision and its evidence are recorded and synchronised to the ground over a secured link.

    Decision recorded, ground informed

Time budget

Every stage inside its budget

Each stage is designed to run inside a fixed time budget, so that the whole chain can answer in real time. The figures depend on the hardware.

On the flight deck

Evidence

  • Vibration trend above the normal bandSource: Live avionics data, in LLiDA
  • Icing conditions in the current phaseSource: Held context, from the avionics
  • High engine vibration procedureSource: Operator QRH, current revision

Possible solutions

  1. Apply the high engine vibration procedureSource: Operator QRH
  2. Reduce thrust on engine 2 within limits and monitorSource: FCOM limits, live data
  3. Review diversion options if the trend continuesSource: Operations manual

Decision

Chosen by the crew. Recorded with its evidence and sent to maintenance control.

Illustrative use case with invented values. It shows how the technologies combine, as an example for discussion.

What AeroBrain is

Four technologies. One governed stack, from the processor to the decision.

Each technology does one job and can be used on its own. Combined, they take real-time data from the aircraft's avionics, read it against approved manuals and records, and produce an advisory that a person can check, with every step recorded.

Every layer is designed to the same requirements: high assurance, auditability and real-time operation inside a fixed time budget. The whole set was developed in-house by SkyDynamics and grew out of long-running research into AI for safety and mission-critical fields.

From the processor to the decisionAnimated diagram: a processor reads live data from the aircraft's avionics; the four AeroBrain technologies (LLiDA, the Contextual Composition Protocol, the RAG engine and the Assurance Small Language Model) work on it one after another; an instructor, crew member or engineer sees the options and their evidence, and takes the decision, which is recorded.

  1. Processor

    RISC-V, ARM and others

  2. Live data

    From the aircraft's avionics

  3. AeroBrain

    Four technologies, one after another

    • LLiDA
    • Protocol
    • RAG
    • Model

  4. Instructor, crew or engineer

    Sees the options and their evidence

  5. Decision

    Taken by a person, and recorded

The stack, from the decision down to the hardware

  • Developed in-house

    Every technology in the set is designed and written by SkyDynamics.

  • Grown out of research

    The set grew out of long-running research into AI for safety and mission-critical fields.

  • Built to be taken into products

    Each technology is designed to be taken into next-generation avionics and other assured products, at the assurance level the product requires.

Research at SkyDynamics

Secured RAG engine

Retrieval you can replay. A secured RAG engine written in Rust, for real-time and edge AI.

RAG (retrieval-augmented generation) grounds an AI answer in sources you approve. In AeroBrain those sources are not only manuals and records: the engine also retrieves over real-time data from the aircraft's avionics, held as time series in LLiDA. An answer can cite a live reading and the procedure that applies to it.

The engine is deterministic: the same question, on the same document snapshot and the same recorded data, returns the same evidence, in the same order, with its source and time. An auditor can replay an answer, not just read it.

The same question, on the same sources, returns the same evidenceIllustrative

  1. The question

    Asked by a person, or raised by an event.

    High engine vibration in icing conditions

  2. Two kinds of source

    Live data from the avionics and your approved documents, each identified.

    Live avionics data
    Time series in LLiDA, time-stamped
    Approved documents
    Manuals and records, one revision set

  3. Ordered fusion and ranking

    Candidates are fused and ranked in a defined order. Nothing is left to chance.

  4. The evidence set

    Readings and passages, each with its source, time or revision.

    • LLiDAEngine 2 vibration: live trend, time-stamped
    • QRHAbnormal procedures: high engine vibration
    • FCOMPower plant: limitations
    • MELPower plant: dispatch conditions

Audit replay

First run: evidence fingerprint
7c41 9be2 05af
Replayed later: evidence fingerprint
7c41 9be2 05af

Identical evidence set

  • Live avionics data and documents

    Retrieves over real-time data from the aircraft's avionics, held in LLiDA, together with approved manuals and records. Each piece of evidence keeps its source, and its time or revision.

  • Written in Rust

    The engine is written in Rust for memory safety without a garbage collector. Safety-critical components use Ada SPARK and C written to MISRA C.

  • Deterministic paths

    Ordered fusion, reproducible ranking and explicit provenance, so the same query, corpus snapshot and recorded data yield the same evidence set for audit replay.

  • Edge, on-premises and air-gapped

    Runs on the device, behind your perimeter or in a disconnected enclave. Inference and corpora stay under your data residency and access policies.

  • A fixed time budget

    Designed to answer inside a fixed time budget. On our reference stack, retrieval through ranking is ~12 ms-class; with LPU-class accelerators the target is ~3 ms-class. Your figure depends on hardware, index size and network path.

  • Operator-grade observability

    Usage, model routing and safety classifications are exposed for enterprise monitoring, so the engine can be operated and evidenced like any other critical system.

Contextual Composition Protocol, developed by SkyDynamics

Context held, not rebuilt. The Contextual Composition Protocol keeps the whole task in memory.

The Contextual Composition Protocol holds all the data relevant to a specific task or workflow in memory, behind the scenes, and correlates it with whatever arrives at that moment: a reading, a question, a message.

So when an event occurs, nothing has to be looked up or entered again. The system already holds the phase of flight and the aircraft state from the avionics, and the procedure in progress, and the event is read in that context.

In service today: EBT and CBTA training

The protocol already works behind every aeroEBT application.

  • In the EBT Scenario Builder it holds the specification, the competency framework and the aircraft type while a scenario is composed
  • In the instructor app it holds the scenario and the event in play during recording and assessment
  • It supports the instructor and never makes the grading decision
See the protocol in aeroEBT

Next: real-time decision support

The same mechanism carries into operations and avionics.

  • An alert is read against the phase of flight and the live aircraft state
  • Retrieval is narrowed to the procedures that apply to that situation
  • The advisory is short, because the context is already known

The protocol does not decide. It makes sure that whatever is decided, by a person or a model, is decided with the full context in view.

Low Level integrated Database Architecture, developed by SkyDynamics

LLiDA. A real-time time-series database for RAM, disk and the processor itself.

LLiDA captures real-time data, such as the data of an aircraft's avionics, as time series. It works in RAM, on disk and at processor level, on RISC-V, ARM and other architectures, from a low-level microprocessor computer to a server.

It keeps every connected node synchronised live, over secured or common networks, and it is written to safety-critical rules such as MISRA C and the objectives of DO-178C, for high-reliability and high-assurance systems.

  • Topproduct

  • Mostinnovative

  • Innovationtech

  • Disruptivetech

  • Highassurance

From the avionics to every node, liveIllustrative

Sampled at a fixed rate

Each avionics channel is read on a fixed period, so timing is known in advance.

Held in RAM and on disk

The newest samples sit in memory for the layers above, and the record is kept on disk.

Latest sample
4128

Synchronised live

Every node sees the same record, as it is written.

  1. Microcontroller

    RISC-V, low-level computer

    4128In sync

  2. Secured network

    Avionics computer

    ARM, on the aircraft

    4128In sync

  3. Common network

    Ground system

    Server, operations and maintenance

    4128In sync

  • Real-time time series

    Avionics and sensor values are captured as they are produced and are available to the layers above without a round trip to a server.

  • RAM, disk and processor

    LLiDA works in RAM, on disk and at processor level, so one architecture covers the working data and the stored record.

  • RISC-V, ARM and others

    The same architecture across processor families, from low-level microprocessor computers to avionics computers and servers.

  • Live synchronisation

    Nodes exchange data live over secured networks or common networks, so the aircraft, the edge device and the ground hold the same record.

  • High reliability and assurance

    Developed to the rules used for flight software, including MISRA C and the objectives of DO-178C, for high-reliability and high-assurance systems.

  • No single point of failure

    Every node holds the record. A node that stops is isolated and the others keep serving the data.

  • Device plugins for iPads

    Crew and engineer iPads read the on-board values through a LLiDA plugin, and the function on board never depends on the device.

  • Cryptographic identity

    Platforms connect only by proving their identity with a key. A device that cannot is refused at the boundary.

  • Vector and semantic search

    Plugins that search readings and documents by meaning on the device, designed for high speed, low latency and low energy.

From the avionics to the iPad

Choose a feature to see it work, from the sensor on the aircraft to the device in a crew member's hands.

From the avionics to the iPad, in real timeIllustrative

Aircraft avionics

Safety-critical, on board

  • Engine 2 vibration0.62 ips
  • Exhaust gas temperature624 °C
  • Hydraulic pressure3011 psi

Sampled on a fixed period

LLiDA nodes

Each holds the record

  • Avionics computerARM
    In sync#8224
  • Edge unitRISC-V
    In sync#8224
  • Ground serverServer
    In sync#8224

Cryptographic boundary

Identity proven by key

  • Crew iPadKey verified
  • Engineer iPadKey verified
  • Unknown deviceRefused: no key

User devices

Through the LLiDA plugin

Crew iPadFlight deck

LLiDA plugin

Engine 2 vibration0.62 ips

Semantic search, on the iPad

engine vibration rising in icing

  1. Engine 2 vibration, last 10 minutes0.94
    Live time series
  2. High engine vibration procedure0.89
    Operator QRH
  3. Engine anti-ice operation0.81
    FCOM

On the device · Low latency · Low energy

Engineer iPadMaintenance

LLiDA plugin

Engine 2 vibration0.62 ips

Every value with its time, on board.

Avionics channels are sampled on a fixed period on the aircraft and held as time series in RAM and on disk, so the layers above read them without a round trip to a server.

LLiDA in full: architecture, plugins, security and fieldsWhat a safety-critical database must do

Assurance Small Language Model

A small model you can audit. The Assurance Small Language Model, designed for real-time use in safety-critical work.

Large general-purpose models are hard to bound and hard to inspect. The Assurance Small Language Model is custom developed by SkyDynamics to be the opposite: compact, so that it can run at the edge in real time, and designed so that each statement it produces can be traced to the evidence it was given.

Every statement keeps its source; what has no source is withheldIllustrative

Evidence in

  • E1Live reading: vibration trend above the normal band
  • E2Held context: icing conditions in the current phase
  • E3Manual: high engine vibration procedure

Assurance Small Language Model

Compact. Fixed version. Runs on the device.

Statements out

  • Engine 2 vibration is rising.Source E1
  • Icing conditions are present.Source E2
  • The high engine vibration procedure applies.Source E3
  • A statement about the probable cause.Withheld: no supporting evidence

Audit record

Input
Event and held context
Evidence
E1, E2, E3 with time or revision
Model
Version identified
Output
Three statements, one withheld

  • Auditable

    Input, evidence, model version and output are recorded together, so a result can be examined after the fact.

  • High assurance

    Developed under the same assurance discipline as the rest of the set, for work where an unsupported answer is not acceptable.

  • Real time at the edge

    Designed to run on the device, without a connection, and to answer inside a fixed time budget. The figure depends on the hardware and is set with the customer.

Why a small model, not a general-purpose LLM

General-purpose LLMLarge modelIn a data centre, over a networkAnswers from its trainingMay state what no source supportsVersion may changeTiming variesAssurance Small Language ModelE1E2E3Evidence inSmall modelOn the deviceFixed versionVibration is risingE1The procedure appliesE2Within limitsE3A probable cause: withheld, no sourceInside the time budget
Why a small model, not a general-purpose LLM
RunsGeneral-purpose LLM: In a data centre, over a networkAssurance Small Language Model: On the device, on premises or air-gapped
Answers fromGeneral-purpose LLM: Its training, plus what it is givenAssurance Small Language Model: The evidence retrieved for the question, only
Without a sourceGeneral-purpose LLM: May still state somethingAssurance Small Language Model: Withholds the statement
VersionGeneral-purpose LLM: May change with the providerAssurance Small Language Model: Fixed, and recorded with each answer
TimingGeneral-purpose LLM: Varies with load and the networkAssurance Small Language Model: Designed for a fixed time budget

A general comparison of the two kinds of model; individual deployments differ.

Safety-critical LLM or SLM, compared

Standards, assurance and auditability

Built to the rules of flight software. Assurance and auditability are designed in, not added afterwards.

The technologies are developed to high-assurance practice: MISRA C for C code, Rust for memory safety, Ada SPARK where proof is required, and the objectives of DO-178C for airborne software. Requirements, code, tests and evidence are traceable in both directions.

Traceability, in both directions

  1. Requirement

    Each function starts as an identified requirement.

  2. Design

    The design states how the requirement is met.

  3. Code

    Written to MISRA C, in Rust or in Ada SPARK.

  4. Test

    Each requirement has tests that exercise it.

  5. Evidence

    Results are kept with the release they belong to.

Forward: every requirement reaches a test.

Back: every line of code answers a requirement.

Auditable by design

High assurance needs more than careful development. Every answer the system gives in service must be open to examination afterwards, by the operator, the manufacturer or the authority.

The answer, recordedInputEvent and its timeEvidenceSource and revisionModelVersion identifiedOutputStatements, one withheldDecisionAccepted by the crewEvidence fingerprint7c41 9be2 05afReplaySame dataReplayed by an auditorInputEvent and its timeEvidenceSource and revisionModelVersion identifiedOutputStatements, one withheldDecisionAccepted by the crewEvidence fingerprint7c41 9be2 05afIdenticalevidence setIllustrative record. Retrieval is deterministic, so the same question on the same data returns the same evidence.

  • Recorded

    The input data and its time, the evidence with its source and revision, the model version, the output and the human decision are kept together.

  • Replayable

    Retrieval is deterministic, so the same question on the same data returns the same evidence when an auditor runs it again.

  • Traceable

    Each statement the model produces is designed to point to the evidence behind it, and each requirement to its code and tests.

  • Accountable

    A person takes the decision. What was proposed, accepted or overridden is recorded with it.

Standards that frame the engineering

MISRA C and the objectives of DO-178C are followed in development. The related standards below frame how the technologies are taken into an assured or certified product with the customer.

Languages and coding rules

  • MISRA CCoding rules for C in critical systemsHere: C code in the set, including LLiDA
  • RustMemory and thread safety enforced by the compilerHere: The RAG engine
  • Ada SPARKA subset of Ada whose properties can be provedHere: Safety-critical components

Airborne and ground software

  • DO-178C / ED-12CSoftware considerations in airborne systems and equipment certificationHere: The development objectives the engineering follows
  • DO-330 / ED-215Software tool qualificationHere: Tools whose output is relied on without further verification
  • DO-333 / ED-216Formal methods supplement to DO-178CHere: Where SPARK proof stands in for testing
  • DO-278A / ED-109ASoftware for ground-based CNS/ATM systemsHere: Ground systems the aircraft synchronises with

Systems, hardware and security

  • ARP4754B / ED-79BDevelopment of civil aircraft and systemsHere: Allocating functions and assurance levels with the integrator
  • ARP4761A / ED-135Safety assessment processHere: The failure conditions that set the assurance level
  • DO-254 / ED-80Design assurance for airborne electronic hardwareHere: The hardware that hosts the software
  • DO-326A / ED-202AAirworthiness security processHere: Synchronisation over secured and common networks

AI and other safety-critical fields

  • EASA AI Roadmap and concept paperGuidance for Level 1 and 2 machine learning applicationsHere: AI trustworthiness, explainability and human oversight
  • EU AI ActRegulation (EU) 2024/1689 on artificial intelligenceHere: High-risk AI obligations
  • IEC 62304Medical device software life cycleHere: Medical applications
  • IEC 61508Functional safety of electronic systemsHere: Other safety-critical industries

AeroBrain is a set of technologies developed to these standards. Certification is granted for a specific product and installation, with the authority, and the engineering is organised so that the evidence can be produced.

Where it applies

Combine what the task needs. For aerospace, defence and other safety-critical fields.

The technologies are specialised in aerospace and defence. They are designed for any field where a wrong answer has consequences, so the same set can be specified for medical and other safety-critical systems. Each can be used alone; the task decides the combination.

How the technologies combine, by fieldIllustrative

  • Next-generation avionics

    Real-time decision support from live avionics data

    LLiDA + Protocol + RAG + Model

  • Operations and maintenance control

    Live fleet data, procedures and recovery options

    LLiDA + Protocol + RAG

  • EBT and CBTA training

    Scenario design, recording and assessment in aeroEBT

    Protocol + RAG

  • Defence and mission systems

    Disconnected and air-gapped operation

    LLiDA + Protocol + RAG + Model

  • Medical and other safety-critical devices

    Monitoring and advice at the point of use

    LLiDA + RAG + Model

The combinations the set is designed for, not a list of deployments. Each deployment is specified with the customer and its assurance level.

Where it stands today

A status is given here only where this site states one. For everything else, ask for the current status in a technical briefing.

Where it stands today
Contextual Composition ProtocolEBT and CBTA training: behind every aeroEBT application, including the EBT Scenario Builder and the instructor appStatus: In service today
Contextual Composition ProtocolReal-time decision support in operations and avionicsStatus: Next
The four technologies combinedReal-time decision support on the aircraft, as in the use case on this pageStatus: Illustrative use case

In SkyDynamics products

aeroEBTComposition Protocol in serviceOpsEyeBuilt on AeroBrainFlying WhaleThe pilot's appAeroTechLogDefect context for maintenanceAeroBrainLLiDA · Protocol · RAG · ModelIn service today: EBT and CBTANext: real-time decision support in avionics
  • aeroEBT and the EBT Scenario Builder

    Evidence-Based Training and CBTA, with the Contextual Composition Protocol behind every application.

  • OpsEye

    Airline operations control: disruption monitoring and costed recovery options.

  • Flying Whale

    The pilot's app: briefing, weather, NOTAMs and logbook.

  • AeroTechLog

    The digital technical log, with defect context for maintenance control.

Deployment and security

Inside your perimeter. On the device, on premises or air-gapped.

High-assurance AI is only as trustworthy as the place it runs and the way it is reached. AeroBrain is designed to run where the data is and to admit only what proves its identity.

Where it runsOn the deviceEdge, no connection neededOn your premisesBehind your perimeterEnclaveAir-gappedNo network at allEUSwitzerlandYour premisesHosted recordsEU · Switzerland · premisesHow it is protectedAdmission by keyCrew iPad: key verifiedUnknown device: refusedProtected in transitNode to node, node to deviceRecorded, to aviation rulesInput, evidence, version, decisionPart-IS alignedDO-326A

Where it runs

  • On the device

    The RAG engine, the Assurance Small Language Model and LLiDA run at the edge, on RISC-V, ARM and other processors, with no connection required.

  • On your premises

    Behind your perimeter, with inference and corpora under your data residency and access policies.

  • Air-gapped

    In a disconnected enclave, for defence and other environments where no network is allowed.

  • Hosted records

    Where records are kept on the ground: in the EU, in Switzerland or on your premises.

How it is protected

  • Admission by key

    Platforms joining LLiDA prove their identity cryptographically; a device without a valid key is refused.

  • Protected in transit

    Data between LLiDA nodes and devices is protected cryptographically, over secured or common networks.

  • Everything recorded

    Input, evidence, model version, output and the human decision are kept together, so any answer can be examined later.

  • Aviation security rules

    SkyDynamics' practices are aligned with EASA Part-IS, and DO-326A frames network synchronisation in an aircraft programme.

Data, security and AI boundaries in the Trust centre

Further reading

Go deeper. Each subject of this page, on its own page.

  • Safety-critical AI

    What makes AI safety-critical, and a safety-critical LLM compared with a small language model.

  • LLiDA in full

    Architecture, plugins, security, the fields it is designed for and a specification sheet.

  • Safety-critical database

    What a database must do for avionics, mission-critical and space operations.

  • AI for aviation and airlines

    Where the AI works today: operations, training, maintenance training and the pilot.

  • Trust centre

    Where data lives, who approves what, and what the AI is not allowed to do.

  • Glossary

    Every term on this page, in one or two sentences.

Questions. Asked by engineers, operators and authorities.

Ask us something else
What is AeroBrain?

AeroBrain is a set of AI technologies developed by SkyDynamics for safety and mission-critical fields, specialised in aerospace and defence. It combines a secured RAG engine written in Rust, the Contextual Composition Protocol, the LLiDA real-time time-series database and an auditable Assurance Small Language Model to turn real-time avionics data and approved documents into decision support that people can verify.

What kind of AI is suitable for aviation?

AI for aviation has to be verifiable. AeroBrain is built for that: retrieval is deterministic, every answer keeps its evidence with its source and revision, the human decision is recorded with it, and the software is developed to safety-critical rules such as MISRA C and the objectives of DO-178C. It is designed to give decision support in real time from avionics data and approved manuals, for safety and mission-critical work in aerospace and defence.

What is LLiDA?

LLiDA (Low Level integrated Database Architecture) is a real-time time-series database developed by SkyDynamics for high-reliability and high-assurance systems. It works in RAM, on disk and at processor level, on RISC-V, ARM and other architectures, synchronises live over secured or common networks, and is developed to safety-critical rules such as MISRA C and the objectives of DO-178C.

What data does AeroBrain work from?

Two kinds. Real-time data from the aircraft's avionics, captured as time series by LLiDA, and the operator's approved manuals and records. The RAG engine retrieves over both, and each piece of evidence keeps its source, and its time or revision.

What is the Contextual Composition Protocol?

The Contextual Composition Protocol is SkyDynamics technology that holds all the data relevant to a specific task or workflow in memory, behind the scenes, and correlates it with each event as it arrives. It already works behind every aeroEBT application, including the EBT Scenario Builder, and carries the same mechanism into real-time decision support.

Does AeroBrain support real-time and hard real-time applications?

It is designed for them. The set is built to work on live data and to answer while the event is still unfolding. For hard real-time functions, where a late answer counts as a failure, each stage is designed to run inside a fixed time budget. The figures depend on the hardware and the function: they are set with the customer and have to be shown for each product.

Is AeroBrain auditable?

Yes. Auditability is part of its high-assurance design. For every answer, the input data and its time, the evidence with its source and revision, the model version, the output and the human decision are recorded together. Retrieval is deterministic, so an auditor can replay the same question on the same data and obtain the same evidence.

How does AeroBrain relate to DO-178C?

The technologies are developed to high-assurance practice, including MISRA C and the objectives of DO-178C, so that certification evidence can be produced. Certification is granted for a specific product and installation, with the authority.

Can AeroBrain run without an internet connection?

Yes. The RAG engine and the Assurance Small Language Model can run on the device or on-premises, including in air-gapped environments, and LLiDA keeps nodes synchronised over whichever secured or common network is available.

What is AeroBrain AI used for today?

The Contextual Composition Protocol is in service today behind every aeroEBT application, including the EBT Scenario Builder and the instructor app. OpsEye, the airline disruption management product, is built on AeroBrain. Real-time decision support in avionics is the next step for the protocol, and the four technologies combined are shown on this page as an illustrative use case.

Does AeroBrain use a large language model (LLM)?

Not as a source of answers. AeroBrain uses the Assurance Small Language Model, a compact model developed by SkyDynamics that composes statements only from the evidence the RAG engine retrieves, withholds what has no source, and runs at a fixed, recorded version on the device.

What is a safety-critical SLM?

A small language model built for safety-critical work: compact enough to run on the device in real time, bounded to its task and auditable. In AeroBrain that is the Assurance Small Language Model.

Which processors and hardware does AeroBrain run on?

RISC-V, ARM and other processors, in RAM and on disk, from a microcontroller to a data centre, connected or air-gapped. The figures depend on the hardware and are set with the customer.

How does AeroBrain handle cyber security?

Platforms joining LLiDA are admitted only by cryptographic identity and data between nodes and devices is protected cryptographically. The engine runs on the device, behind your perimeter or air-gapped, every answer is recorded, and SkyDynamics' practices are aligned with EASA Part-IS.

Can crews and engineers use AeroBrain data on an iPad?

Yes. Through the LLiDA device plugin, crew and engineer iPads read the on-board values across a cryptographic boundary, while the function on board keeps running on its own nodes.

Does AeroBrain support vector and semantic search?

Yes. LLiDA plugins provide vector and semantic search on the device, designed for high speed, low latency and low energy, so evidence can be found by meaning at the edge.

Can each AeroBrain technology be used on its own?

Yes. LLiDA, the Contextual Composition Protocol, the RAG engine and the Assurance Small Language Model can each be used alone; the task decides the combination.

How do we start with AeroBrain?

With a technical briefing: tell SkyDynamics the function, the hardware and the assurance level, and the briefing shows which technologies apply and what evidence comes with them.

Does AeroBrain make the decision?

No. AeroBrain supports the decision. It proposes, shows its evidence and records what was accepted or overridden. The accountable person decides.

Bring high-assurance AI into your programme. One briefing shows which technologies apply and what evidence comes with them.

Tell us the function, the hardware and the assurance level.

Request a technical briefingHow our AI is governed

Up next

Keep exploring