Embedded software intern

Internship and graduation projects Q1-Q2 2027
Student projects Q1-Q2 2027
Your knowledge. A real engineering challenge.
Would you like to work on embedded systems, cybersecurity, AI or the Internet of Things during your internship or graduation project? At Inspiro, you will research, design and build a solution that you can test in practice.
Explore nine internship and graduation projects for the first half of 2027. Are you a bachelor’s or master’s student in a field such as Computer Science, Embedded Systems or Electronics? Find the technical challenge that matches your interests.
Each project combines research with a working prototype. Expand the project goals to explore the technical details.
Embedded software Linux Cybersecurity
Design and implement a reusable test tool that automatically verifies whether embedded systems establish secure connections correctly and reject insecure connections. Investigate the most suitable test approach, for example a Hardware-in-the-Loop (HIL) setup with a configurable TLS test server. The tool should cover both valid connections and failure scenarios, such as certificates signed by an untrusted root CA, expired certificates, or a server name mismatch. Deliver a working prototype and demonstrate the test suite on a representative embedded system.
Project title: Reusable TLS test suite
Test approach and architecture: Evaluate existing tools and the added value of HIL. Design a modular test setup that supports different devices and TLS implementations with minimal adaptation.
Automated TLS test scenarios: Implement repeatable tests for valid connections and failure cases, including an untrusted root CA, expired or not-yet-valid certificates, a server name mismatch, and an invalid certificate chain. Include mutual authentication (mTLS) where relevant.
Reliable result verification: Verify that valid connections succeed and invalid connections are rejected. Check that the system does not fall back to unencrypted communication when errors occur. Report expected and observed behaviour for each test, supported by relevant logs.
Reusability and integration: Make test scenarios, certificates, and device configurations easy to manage. Support automated execution and integration into a CI/CD pipeline or an existing test environment.
Validation and handover: Validate the tool using deliberately introduced configuration errors and demonstrate its operation on an embedded system. Deliver documentation for installation, use, and extension, including test coverage and limitations.

Cybersecurity Embedded Linux CI/CD
Design and implement an automated pipeline that collects security evidence for each firmware release. Connect information about software components, known vulnerabilities, build provenance, and security test results to the exact firmware artifact delivered. Investigate how to identify components actually included in the firmware and reduce irrelevant vulnerability findings. Deliver a working prototype that supports security review and product maintenance, with clear traceability between releases and their supporting evidence.
Project title: Firmware security evidence pipeline
Component inventory: Generate a Software Bill of Materials (SBOM) and evaluate its completeness against the components included in a representative firmware build.
Vulnerability assessment: Match components to known vulnerabilities and provide a documented review process for relevance, uncertainty, and accepted findings.
Evidence collection: Associate source revisions, build configuration, artifact hashes, and security test results with each release.
Release reporting: Produce a readable security report that highlights changes, unresolved findings, and missing evidence.
Validation and handover: Evaluate the pipeline using deliberately introduced vulnerable components and incomplete evidence. Document its coverage and limitations as a tool supporting compliance activities.

Embedded software AI Signal processing
Edge AI for machine condition monitoring
Design and implement an embedded monitoring system that detects abnormal machine behaviour using vibration, sound, or electrical current measurements. Run detection locally on a microcontroller and investigate the trade-offs between compact machine learning models and conventional signal processing. Deliver a working prototype for one representative machine, such as a motor, pump, or fan, and evaluate it using reproducible operating conditions and fault scenarios.
Project title: Edge AI for machine condition monitoring
Measurement setup and dataset: Select suitable sensors and collect representative data for normal operation and selected abnormal conditions. Separate training and evaluation data to assess performance on unseen measurements.
Detection approach: Compare a lightweight machine learning model with a conventional signal-processing baseline, evaluating detection accuracy, false alarms, and robustness to changing operating conditions.
Embedded implementation: Deploy the selected approach on a microcontroller and measure memory usage, processing latency, and energy consumption.
Validation and handover: Demonstrate real-time detection and document the implementation, measured performance, limitations, and steps required to adapt it to another machine.

Software & hardware Embedded FPGA Production automation
Design and implement a configurable production station that combines firmware programming with functional testing of embedded products. The station should identify the connected product, select the appropriate firmware and test sequence, program the device, and verify its basic functionality. Investigate how shared programming and test logic can be separated from product-specific fixtures and interfaces. Deliver a working prototype supporting two representative product variants, with clear operator feedback and traceable production results.
Project title: Universal Functional production programmer
Modular station architecture: Design a reusable framework with adapters for programming tools, test equipment, and product-specific fixtures.
Product identification and programming: Verify product identity and hardware revision before selecting, programming, and checking the approved firmware.
Functional verification: Implement configurable checks for relevant functions, such as power consumption, digital and analogue I/O, sensors, and communication interfaces.
Operator workflow and traceability: Provide clear progress and pass/fail feedback. Record the device identifier, hardware revision, firmware version, test results, and station configuration.
Validation and extensibility: Demonstrate operation on two product variants and evaluate programming failures, incorrect product selection, and interrupted test sequences. Document how to add another product.

Embedded software Linux EV charging
Open source charge point (EVerest)
Design and implement a Linux + EVerest EV charging station controller (AC, Multi-connector) that connects securely to a CSMS using OCPP2.0.1 (JSON over WebSockets + TLS/mTLS). Deliver a working prototype, plus a structured compatibility investigation for ISO15118 (AC), covering standards, Plug & Charge (PnC) PKI, interfaces, simulators, and the implications for a future integration
Project title: Open source charge point (EVerest)
Prototype EVSE controller on Linux/EVerest: Hardware platform selection, contactor control, metering integration (adapter), basic HMI/logging, status handling.
OCPP 2.0.1 client (minimum subset) to a CSMS simulator, including device management, transaction handling, metering, firmware/log flows, and security profile with TLS/mTLS and correct time/cert validation.
Maintenance dashboard: Basic dashboard with charge point maintenance information like current state of operation.
ISO 15118 compatibility investigation: Produce a gap analysis vs. the prototype and a migration plan for later implementation.

Embedded software Cellular IoT Cybersecurity
Resilient remote connectivity management for IoT devices
Design and implement a connectivity management solution that allows remotely deployed IoT devices to change their cellular connectivity configuration without physical access. Use an IoT eSIM provisioning solution based on SGP.32 and investigate how devices can recover when a profile change fails or the new connection cannot reach the application backend. Deliver a working prototype on a representative embedded platform. Demonstrate remote profile management, connection verification, and recovery under controlled failure scenarios, using suitable eSIM hardware and an available provisioning service.
Project title: Resilient remote connectivity management for IoT devices
Architecture and technology selection: Investigate the roles of the device, eSIM, provisioning service, and application backend. Select a suitable platform and define which management and recovery functions each component supports.
Remote profile management: Implement an authenticated workflow for requesting and tracking profile changes. Verify successful connectivity at application level before reporting the operation as complete.
Resilient recovery: Design a recovery state machine for interrupted provisioning, unavailable networks, and unreachable backend services. Evaluate when local recovery or fallback to a previously working profile is possible.
Automated failure testing: Introduce power interruptions, network loss, and backend outages at selected stages. Measure recovery time, energy consumption, and the number of scenarios requiring manual intervention.
Validation and handover: Demonstrate supported profile changes and recovery scenarios. Document integration requirements, service dependencies, security considerations, and limitations.

Embedded software Wireless connectivity Low-power systems
Delay-tolerant IoT sensing over cellular and satellite networks
Design and implement an IoT sensor node that reliably collects and delivers measurements when network connectivity is intermittent. Use cellular connectivity when available and investigate satellite IoT as an alternative for locations outside terrestrial coverage. Develop a store-and-forward mechanism that preserves measurements during outages and resumes delivery when connectivity returns. Deliver a working prototype and evaluate the trade-offs between delivery reliability, latency, energy consumption, and communication cost through controlled tests and an available satellite connection.
Project title: Delay-tolerant IoT sensing over cellular and satellite networks
Platform and connectivity assessment: Select a suitable sensor, embedded platform, and cellular/satellite service combination. Determine coverage constraints, supported payload sizes, and whether switching requires modem reconfiguration or restart.
Persistent data buffering: Implement a bounded measurement queue that survives power loss. Define timestamp handling, message identifiers, delivery acknowledgements, duplicate detection, and behaviour when storage becomes full.
Adaptive transmission strategy: Develop a configurable policy for network selection, batching, retries, and message priority. Balance delivery deadlines against energy use and communication cost.
End-to-end validation: Build a receiving application and test prolonged outages, interrupted transmissions, duplicate deliveries, and unexpected resets. Measure data completeness, delivery latency, and energy per delivered measurement.
Demonstration and handover: Demonstrate cellular delivery and satellite transmission where service access permits, complemented by repeatable outage simulations. Document measured performance, operating limits, and recommendations for deployment.

Embedded software Low-power electronics Test automation
Design and implement an automated test setup that detects increases in energy consumption between firmware versions. Measure representative operating scenarios, such as sensing, wireless communication, and sleep cycles, and relate the measurements to device behavior. Investigate how to distinguish meaningful regressions from normal measurement variation. Deliver a prototype that makes energy consumption a measurable part of firmware quality assurance.
Project title: Energy-aware firmware regression testing
Measurement setup: Select and validate equipment capable of measuring both low-power sleep states and short current peaks.
Repeatable test scenarios: Automate representative device activities and synchronize power measurements with relevant firmware events.
Regression detection: Establish baselines and acceptance thresholds that account for measurement noise and variation between test runs.
Automated reporting: Compare firmware versions using energy per operation, sleep current, and time spent in different operating states.
Validation and integration: Demonstrate detection of deliberately introduced energy regressions and integrate the setup into an existing test workflow or CI pipeline.

AI Embedded systems Test automation
Investigate how a language model can support the generation of embedded software tests from requirements and interface descriptions. Develop a prototype that proposes test cases, converts selected cases into executable tests, and runs them in an existing test environment. Evaluate whether the approach identifies additional defects and reduces engineering effort compared with a manually created baseline. Focus on one software module or communication interface, with independently defined expected behavior.
Project title: AI-assisted embedded test generation
Requirements-based test generation: Generate traceable test proposals covering normal behaviour, boundary conditions, and invalid inputs.
Executable test integration: Convert reviewed proposals into tests compatible with the selected framework and target environment.
Independent result verification: Derive expected outcomes from reviewed requirements or a trusted reference implementation, and identify ambiguous requirements for human review.
Quality and effort evaluation: Compare generated and manual tests using deliberately introduced defects, false alarms, reproducibility, and the time needed to review and correct tests.
Validation and handover: Deliver a working prototype and document where AI assistance adds value, where it fails, and which review steps remain necessary.

Learn, develop and join in.
At Inspiro, you will work in an environment with a passion for engineering and innovative solutions. During your internship, an experienced engineer will guide you as your buddy.
Guidance from engineers with in-depth technical expertise.
An open learning environment with room to develop your skills.
An informal atmosphere where you are part of the team.
Join our Friday afternoon drinks and other company activities.

Ready for your next step?
Found a project that suits you? Tell us which project interests you, what you are studying and when you would like to start. You can get in touch directly through our contact page.
Inspiro Arnhem Internship and graduation projects Q1-Q2 2027
Are you looking for a partner in technology or do you have a specific challenge? Our team of experts is ready to help you.
Have a specialist call you back
Have a specialist call you back
Leave your details and we will contact you personally.
Send a message for personal contact
Leave your details and we will contact you without obligation.