Job Description

Want to be part of transforming road freight? Einride is showing the world a new way to move, based on the latest digital, autonomous and electric technologies. We build the technology that hauls freight efficiently, and we put it to work for the world's leading manufacturers, retailers, logistics providers and defense organizations, all working toward a future where moving goods cost less for our customers, and less for the planet.

Founded in 2016, Einride became the first company in the world to deploy a cab-less autonomous electric vehicle on a public road (Sweden, 2019). In 2022, we were the first to successfully operate such a vehicle on a U.S. public road. In June 2026, Einride became a publicly listed company on Nasdaq, a milestone that reflects a decade of turning what once looked like science fiction into everyday reality. Today, our technology moves freight across North America, Europe, and the Middle East for some of the world's biggest shippers.

As a Master’s Thesis student at Einride, you’ll be joining a top talent team of ambitious, creative, kind-hearted people that’s pioneering a new era for road freight. We think big, and aim for zero emissions.

Einride cultivates close relationships with academia and welcomes a limited number of Master’s students to write their thesis with us each year. During the thesis period, you will be supervised by one of our team members that are equally as passionate about your subject area as yourself. Together, you will throughout the semester work on projects that will make a real difference toward a sustainable future.

Requirements:

• Currently finishing your Master’s degree and writing your thesis during Spring 2027
• A highly driven individual who thrives working as part of a collaborative team.
• A strong interest within the tech and startup scene.
• A passion for sustainability.

About the recruitment process and program

We aim for the recruitment process to be completed in December. We recommend submitting it as soon as possible, as selection and interviews will be held continually.

The Master's Thesis Program runs from January 18th to June 4th, 2027. Our offices are located in Stockholm and Gothenburg, and while we highly encourage on-site collaboration for optimal results, we are fully set up to welcome applications from students working remotely from any university in Sweden. All of our Master’s Thesis projects are remunerated upon completion.

Below you'll find the research areas for this upcoming Master's Thesis Program:

Central Functions - Public Affairs: Diesel Trucks as Stranded Assets: Residual Value Risk and the Limits of Financial Regulation

Evaluate residual value risks and stranded-asset exposure for diesel trucks during the electric transition to assess whether current banking frameworks and financial regulations adequately price emerging market risks.

Central Functions / Autonomous Technologies - Connectivity + Verification & Validation: AI-Driven Connectivity Resilience and Safety-Aware Handover Validation for Autonomous Trucks

Ensure uninterrupted autonomous truck operations in complex logistics environments by developing AI-driven connectivity management—predicting network degradation and enabling proactive handovers between 5G, satellite, and Wi-Fi networks to prevent unnecessary safety stops.

Electric Mobility - Solutions / Data Science: Risk-Adjusted Lane Pricing for Electric Freight Fleets under Uncertain Utilization, Charging, and Backhaul

Optimize freight quoting for electric fleets under real-world uncertainty by developing a probabilistic pricing framework—balancing win probability against downside risk across unpredictable charging, utilization, and backhaul scenarios.

Electric Mobility - Operations: Can Electric Trucks Serve Short-Notice Freight? Feasibility, Loadboard Access and Cost Compared with Diesel

Maximize fleet utilization for electric trucks by evaluating opportunities in short-notice freight and loadboards—analyzing charging logistics, route constraints, and TCO directly against traditional diesel operations.

Electric Mobility - Operations: Evaluating Operational Cost Parity in Heavy Electric Freight: Technology Maturity, Tipping Points, and OEM Differences

Identify the tipping points for electric freight profitability by building an empirical TCO model—evaluating real-world maintenance costs, OEM performance variations, and key operational levers against traditional diesel benchmarks.

Energy & Charging Infrastructure - Energy Solutions: Designing an eFloater: A Standardized Price Index for Charging Battery-Electric Heavy-Duty Trucks in Europe

Enable fair energy risk-sharing and accelerate electric transport adoption by developing a standardized European charging price index ("eFloater")—transparently indexing real-world electricity, grid, and operational costs for contractually robust freight pricing.

Autonomous Technologies - Motion Planning & Control: Optimization on Graphs of Convex Sets for Behavior Planning in Autonomous Driving

Integrating Graphs of Convex Sets (GCS) into motion planning enables globally optimal decision-making by unifying discrete decisions, such as traffic rules, right-of-way, and lane choices, with continuous trajectory optimization to tackle an underexplored frontier in autonomous driving.

Autonomous Technologies - Motion Planning & Control: Safe Deployment of VLA Models in Autonomous Driving

Bridge the gap between advanced AI and real-world safety by wrapping slow, unpredictable Vision-Language-Action (VLA) models in runtime monitors and Einride’s deterministic safety shield—allowing autonomous vehicles to safely use smart foundation models on dynamic roads.

Autonomous Technologies - Motion Planning & Control: Vehicle System Identification and Physics-Informed Simulation

Elevate simulation fidelity by combining first-principles vehicle physics with Physics-Informed Neural Networks (PINNs) trained on real driving data—capturing complex friction and tire dynamics to build accurate simulation models for autonomous driving.

Autonomous Technologies - Verification & Validation: Data-Driven Scenario Reconstruction and Generation for Safety Validation of Autonomous Trucks

Accelerate safety validation for autonomous trucks by transforming real-world driving data into high-fidelity, parameterized simulation scenarios—systematically generating rare, safety-critical variations to test ODD boundaries and fallback strategies beyond physical track limits.

Autonomous Technologies - Perception: Robust Perception in Adverse Weather Conditions for Autonomous Trucks

Expand operational design domains (ODDs) for autonomous heavy-duty trucks by developing resilient perception pipelines that mitigate failure modes caused by severe weather and poor lighting, enabling safe all-weather deployment.

Autonomous Technologies - Perception: Diagnosing Sensor Degradation in Autonomous Trucks with Tool-Using LLM/VLM Agents

Maximize fleet uptime and safety in harsh winter conditions by developing tool-using LLM/VLM agents to accurately diagnose sensor degradation—distinguishing severe weather from true hardware faults to prevent unnecessary downtime and unsafe operations.

Autonomous Technologies - Perception: VLM-Based Auto-Labeling of Lidar and Camera Data for 3D Perception

A VLM-driven auto-labeling model that lifts open-vocabulary semantics into lidar tracks, yielding accurate, time-consistent 3D pseudo-labels for scalable perception training with minimal human annotation.

Autonomous Technologies - Vehicle Platform & Propulsion: A Systems Engineering Framework for Dynamic Actuator Capability Reporting in Heavy Autonomous Vehicles

Enable heavy autonomous trucks to drive within what their steering and braking can actually deliver, by predicting and reporting how actuator capability changes over time, and making motion planning and control respect those shifting boundaries instead of relying on fixed, preset capability levels.

Autonomous Technologies - Data & Control Tower: Evaluating Spatial and Scenario-Driven Dataset Curation for Multimodal Computer Vision Networks

Ensure the true real-world reliability of our multimodal perception models by developing intelligent data-splitting frameworks. This research focuses on eliminating spatio-temporal data leakage and strategically stratifying rare, critical edge cases to accurately evaluate zero-shot performance.

Autonomous Technologies - AD Hardware: How can a modern compute unit balance real time requirements and flexibility in sensor ingress interfaces?

Maximize compute efficiency and hardware reconfigurability in autonomous vehicles by designing dynamic task scheduling and temporally isolated sensor ingress architectures—optimizing real-time execution across fluctuating camera, LiDAR, and radar data.

Autonomous Technologies - AD Hardware: Deterministic, Latency-Bounded Data Pipeline for Multi-Camera System Optimisation

Guarantee real-time safety and eliminate frame drops in L4/L5 multi-camera systems by building a deterministic, zero-copy data pipeline—leveraging Direct Memory Access (DMA/RDMA) to bypass CPU overhead and achieve sub-10ms sensor ingestion.

Autonomous Technologies - AD Hardware: Native Color Lidar Sensor evaluation

Shape next-generation autonomous truck architectures by quantifying the system value of native color LiDAR—evaluating how time-aligned RGB and 3D point cloud data improve perception accuracy and simplify sensor setups compared to traditional separate hardware configurations.

Autonomous Technologies - AD Hardware: Deterministic, Hardware-Accelerated Near-Lossless Compression for Safety-Critical Multi-Camera Autonomous Perception

Eliminate memory bus congestion and guarantee real-time Level 4 safety by developing a deterministic, near-lossless video compression engine—delivering fixed execution latency and zero frame-drops without degrading AI perception accuracy.