Industry — Automotive

Engineering for automotive software that has to survive vehicles, fleets, audits, and OTA updates.

Software-related recalls surged 80% in 2024. UNECE R155/R156 cybersecurity certification is now mandatory for EU type approvals. The ADAS market is growing at 21% annually. We help OEM teams, Tier suppliers, and mobility platforms build connected-vehicle cloud systems, diagnostics tools, telemetry pipelines, HMI applications, and ADAS-adjacent software with production constraints in mind from day one.

See relevant work ->

Safety and compliance note

We are an engineering partner, not a vehicle certification body or homologation consultancy. We support automotive programs with architecture, implementation, testing evidence, traceability, cybersecurity-by-design, and documentation that fits your ISO 26262, ISO 21448, ISO/SAE 21434, UN R155, or UN R156 process. Final safety, regulatory, and type-approval responsibility remains with the manufacturer and its certified assessors.

Where we help

The automotive software problems that need disciplined engineering, not generic app delivery.

Automotive software programs fail when engineers treat the vehicle, cloud, safety, and cybersecurity layers as separate concerns to be integrated later. They are not — and the regulatory clock is running.

E/E architecture is moving toward zonal and central compute

A typical mid-size vehicle runs 100+ ECUs today. The SDV roadmap targets 3–5 domain or zone controllers — but migration is a 5–10 year program, not a platform swap. Software that ran isolated in a dedicated ECU must share compute resources on a mixed-criticality platform, requiring hypervisor-based ASIL decomposition and AUTOSAR Adaptive expertise that most teams are still building.

ADAS is table stakes in regulated markets — production validation is the hard part

The EU General Safety Regulation mandated AEB and lane-keeping for all new vehicles from July 2024. The ADAS software market reached $12.1 billion in 2025 at 21% CAGR. The demo is not the hard part. Sensor fusion, HIL test coverage, scenario libraries, distribution shift monitoring, and safety evidence are where programs run into trouble — and where we can help.

Cybersecurity and OTA are now certified product architecture

UNECE R155/R156 has been mandatory for EU, Japan, and Korea type approvals since July 2024. Software-related recalls surged 80% in 2024 — 44% of all recalled vehicles had software root causes. Release pipelines, SBOMs, device identity, rollback, and incident response must be built in from the start. The automotive cybersecurity market is projected at $21.44 billion by 2035 at 18.5% CAGR.

Connected-vehicle data is high-volume, long-lived, and safety-adjacent

More than 55% of new vehicles now support OTA. The OTA market reaches $5.93 billion in 2026 and is projected at $29 billion by 2035. Telemetry arrives with gaps, latency, format drift, and multi-tenant fleet boundaries. Dashboards are not enough — the data layer must support diagnostics, warranty analysis, and product decisions, and must survive a vehicle service life of 15 years.

Cockpit and HMI software has to be usable under pressure

In-car UX is not a consumer web app. EU GSR also requires driver monitoring systems in new vehicles. HMI must respect driver attention, hardware constraints, voice and touch interaction, offline states, localization, and brand requirements — while integrating tightly with vehicle services and safety functions underneath.

Diagnostics and aftersales tools are fragmented and underinvested

Service-center workflows, fault-code interpretation, mobile cockpit data, firmware state, and fleet history often live in separate systems with no unified API. Technicians waste time context-switching between tools. We build the integration and workflow layer that makes those signals actionable — reducing diagnostic time and improving first-fix rates.

25–28% fuel cost reduction for a 500+ vehicle logistics fleet

A logistics operator needed route planning that responded to vehicle constraints, live traffic, weather, and delivery windows. We built an optimization engine with dispatch visibility and mobile route updates using Google OR-Tools and real-time traffic modeling — reducing fuel costs by 25–28% and average delivery time by 30%. The same fleet-scale geospatial and vehicle-constraint engineering patterns apply directly to connected vehicle and mobility programs.

PythonGoogle OR-ToolsPostGISAWS

Read the case

Cloud-agnostic IoT platform with secure OTA update support at 2M sensor events/hour

MirrorIOT needed a vendor-neutral platform for high-throughput sensor ingestion, tenant-isolated device management, and secure firmware updates. We built the platform with MQTT, gRPC streams, device-level mutual TLS, data retention policies, and OTA workflows — processing 2 million sensor events per hour. The architecture directly maps to connected-vehicle programs: device identity, update safety, telemetry quality, and fleet-scale rollback.

GolangMQTTgRPCKubernetes

Read the case

Diagnostics platform covering 12,000+ service centers

An automotive service network needed a unified server-side diagnostics platform to replace fragmented per-center tooling. We built the ingestion layer, fault-code normalization service, and workshop-facing interface — enabling consistent diagnostics workflows and centralized fleet health visibility across 12,000+ service locations.

PythonNode.jsPostgreSQLAWS

Read the case

Real-time telemetry for 800+ vehicles — 22% incident rate reduction in 6 months

A commercial fleet operator needed real-time GPS and CAN bus data from 800+ vehicles across multiple markets, plus a mobile cockpit app for drivers. We built the full telemetry pipeline (MQTT, TimescaleDB), CAN data normalization across 5 vehicle makes, a dispatch dashboard with under-3-second update latency, and a React Native mobile app adopted by 100% of drivers.

PythonMQTTTimescaleDBReact Native

Read the case

100% voice-controlled infotainment — passed OEM integration testing on first submission

An automotive Tier 1 supplier needed an embedded infotainment system with full hands-free voice control across navigation, media, climate, and phone — on a cost-constrained ARM SoC. We built the on-device NLP pipeline with sub-400ms latency using quantized ONNX models, a Qt/QML touch interface certified under driver attention requirements, and CAN bus integration for climate commands.

C++17Qt/QMLONNX RuntimeYocto Linux

Read the case ->

System types

What we build for automotive and mobility teams.

Most engagements start with a Product Pilot or architecture phase — especially when safety, cybersecurity, OTA updates, or hardware integration are involved. We scope the system boundary before writing production code.

Start with a Product Pilot ->

Connected-car cloud platforms and fleet portals
Vehicle telemetry ingestion, normalization, and analytics
Diagnostics, fault-code, and service-center workflow tools
HMI, cockpit, and mobile companion applications
Navigation, routing, and location-intelligence systems
OTA update orchestration and release-management support
ADAS data pipelines, labeling workflows, and validation tooling
Embedded, edge, and vehicle-to-cloud integration layers
Cybersecurity architecture support for UN R155/R156 programs
Automotive data platforms for warranty, quality, and operations

Market context

The vehicle has become a software-defined platform. The engineering consequences are significant.

Three regulatory and market forces are reshaping what automotive software programs need to deliver — and what getting it wrong actually costs.

Software quality

80% more software recalls in 2024. 44% of all recalled vehicles had software root causes.

Software-related recalls surged from 112 cases in 2023 to 202 in 2024 — 15 million US vehicles recalled for software issues in a single year (Envorso/WardsAuto). This is not a temporary spike. It is the engineering gap between the speed of software-defined vehicle adoption and the quality infrastructure to support it. OTA updates fix this at scale — but only when the update architecture itself is sound.

Regulatory mandate

UNECE R155/R156 mandatory since July 2024. CSMS and SUMS certification required for EU type approvals.

R155 mandates a certified Cybersecurity Management System (CSMS) based on ISO/SAE 21434 — covering TARA, requirements traceability, and a permanent PSIRT process. R156 mandates a Software Update Management System (SUMS) with cryptographic signing, delta updates, rollback capability, and campaign orchestration. These are not compliance checkboxes — they shape architecture decisions from day one.

OTA and ADAS growth

OTA market at $5.93B in 2026. ADAS software at $12.1B, growing 21% annually.

More than 55% of new vehicles now support OTA capabilities (GM Insights). The OTA update market is projected to reach $29 billion by 2035 at 19% CAGR — driven by regulatory mandate, the shift to feature-as-a-service models, and the need to remediate software defects without dealer visits. The ADAS software market reached $12.1 billion in 2025 at 21% CAGR, with EU regulation now mandating AEB, lane-keeping, and driver monitoring in all new vehicles.

How we engage

Start narrow enough to prove the architecture, then widen deliberately.

Automotive software programs fail when teams treat vehicle interfaces, OTA rollback, data ownership, service workflows, cybersecurity evidence, and safety boundaries as future integration problems. They surface late and expensively. We scope those constraints early and build a vertical slice that proves the riskiest assumptions before committing to full delivery.

System boundary and risk map

Clarify what touches the vehicle, what stays in cloud or mobile, what is safety-relevant, what is cybersecurity-relevant under R155, and where certified partners or client compliance owners must lead.

Architecture and evidence plan

Define interfaces, data contracts, update paths, logging, traceability, test layers — including HIL regression scope — and handoff artifacts before production implementation starts.

Vertical slice build

Build the smallest useful version across the real stack: device or simulator input, backend services, UI workflow, observability, and regression checks. Working software that proves the architecture, not a prototype.

Scale, validate, and transfer

Harden for volume, edge cases, incident response, and long-term maintenance. Your internal team understands what we built and why — the architecture decisions are documented, not locked in our heads.

Fit

Where Insoftex is a strong fit — and where you need a certified specialist alongside us.

Strong fit

Connected-vehicle backend, data, and cloud platform engineering
Diagnostics, service workflows, and aftersales tooling
ADAS-adjacent data pipelines, validation tooling, and simulation infrastructure
OTA workflow, release governance, and device or fleet management software
HMI and cockpit applications, mobile companion apps, and navigation systems
Cybersecurity architecture design and documentation for R155/ISO 21434 programs
Senior engineering capacity for product teams with automotive domain owners in place

Bring a certified specialist too

Vehicle type approval, homologation, or regulatory submission ownership
ASIL determination, final ISO 26262 safety sign-off, and safety assessment
Formal penetration testing or cybersecurity certification as the sole deliverable
Production control software where Insoftex does not own the safety case with your team
Claims of full L4/L5 autonomy without a defined operational design domain and validation strategy

Useful reading for automotive engineering teams.

Insight Software-Defined Vehicles in 2026

OTA architecture, AUTOSAR Adaptive, R155/R156 SUMS/CSMS requirements, and the SDV transition in production programs.

Insight Building ADAS Software in 2026

Sensor fusion architecture, SAE levels, ISO 26262 safety certification, and production validation constraints.

Insight What Is ADAS?

A clear technical primer for teams aligning product scope and engineering requirements.

Service Data Engineering

Telemetry pipelines, event streams, time-series storage, and analytics foundations for connected vehicle programs.

Research basis

Standards and sources this page is aligned with.

Market figures and regulatory framing are grounded in primary sources. We translated them into practical engineering scope — not marketing claims about autonomy.

Client feedback

What clients say about working with us.

We brought Insoftex in after our second failed attempt at productionising the model. In six weeks they rebuilt the inference layer, instrumented it properly, and gave us an eval harness our own team could extend. They told us no twice during the engagement — both times they were right.

Jonathan Langley

CTO · Azarc · UK

We're very happy with the outcome and already looking ahead to the next phase. What stood out most was Insoftex's strong sense of ownership, transparent and fast communication, and ability to think beyond the initial scope to continuously add value. From day one, they supported us in shaping the product vision, through to delivering a high-quality MVP. The result is a robust platform that enables customers to easily book advertising placements and effectively drive visibility and sales. A reliable and forward-thinking partner.

Fei Cheong

General Manager · US

We had a great experience working with the Insoftex team. They played an important role in delivering a modern application for power quality and energy generation analytics, owning both front-end development and automated QA. They built a flexible, user-centric dashboard with configurable widgets, making it easy to analyze data across devices, parameters, and time ranges. Insoftex combines strong technical expertise with a clear focus on quality and delivery. A reliable partner I'd confidently recommend.

Shimon Yannay

Head of Software Development · Israel

Collaborating with Insoftex on our healthcare project proved to be transformative. Their team skillfully re-architected our platform based on comprehensive feedback, delivering exceptional results. They effectively addressed complex challenges while maintaining a strong emphasis on quality and precision. We look forward to continuing our partnership and highly recommend Insoftex to anyone seeking innovative, high-quality solutions.

Dmitry Shteyn

CTO · VURVhealth · USA

Working with Insoftex on the social engagement platform, The Club of Names, was both productive and inspiring. They were involved far beyond development — they helped shape the product's concept and actively contributed ideas that strengthened its core functionality. Together, we built a platform that provides information about names, generates personalized articles, helps users select baby names, and includes a social feature — a chat for people with the same name to connect.

Jason Walker

CEO · JWALKER Marketing · USA

I am happy to share my experience with Insoftex. They made for us a custom .NET application, and it is working very well! It fits perfectly with our needs, and the team did an excellent job integrating it within our local network. Their communication with Azure was seamless, and their professionalism made a big difference. We are pleased with the result and can highly recommend Insoftex for their dedication to quality work!

Thomas Marquardt

CEO · Marquardt Informatik · Germany

We are delighted to acknowledge that Insoftex skillfully programmed our frontend using React, meticulously bringing our design to life. Their adherence to our timelines and effective communication ensured a seamless and productive collaboration.

Ingmar Kruse

CEO · Sun Sniffer · Germany

This has been an amazing experience working with Insoftex, between the communication, the collaboration, and commitment to delivering results it has exceeded our hopes.

Madison Pratt

CTO · DLTChain · Canada

They don't do standard, off-the-shelf products. Rather, they keep their eyes on the market for the newest trends.

Chad Taylor

CEO · Hudson INC · USA

Insoftex's work quality surpassed my expectations. You are fantastic partners.

Emmie Reese

CEO · EpicFlow · USA

Insoftex team have been professional and enthusiastic. The team was always available (even during US-hours). Great job!

Andrew Wilson

CTO · Stealth Startup · USA

Building automotive software with real production constraints?

Book a 30-minute technical call. Bring the system boundary, safety assumptions, update model, data flows, and what has to be true before your team can ship.

Book a 30-min technical call

A senior engineer replies within one business day, often faster.