Software Engineering Intern (Dispatch – Fleet Optimization)

Who we are:

Glydways is reimagining what public transit can be. We believe that mobility is the gateway to opportunity—connecting people to housing, education, employment, commerce, and care. By making transportation more accessible, affordable, and sustainable, we empower communities to thrive and unlock economic and social prosperity.

Our mission is to revolutionize transit with a solution that delivers high capacity, exceptional user experiences, unmatched affordability, and minimal environmental impact.

The Glydways system is a groundbreaking network of carbon-neutral, interconnected transit pathways powered by standardized autonomous vehicles on dedicated roadways. Operating 24/7 with on-demand access, it offers personalized and efficient mobility—without the burden of heavy upfront infrastructure costs or ongoing taxpayer subsidies.

With Glydways, we’re building more than a transportation system; we’re creating a future where everyone, everywhere, has the freedom to move.

Meet the team:

The Dispatch team is responsible for how the Glydways fleet moves as a coordinated system. We design and implement algorithms for:

• Fleet-level planning and optimization (e.g., load-balancing vehicles across stations, choosing when and where to charge)
• Real-time routing and deconfliction to ensure vehicles move safely and efficiently through various zones of a transit system.
• Large-scale simulations used to size fleets, evaluate new system designs, and validate new algorithms before they reach real vehicles

The team is a mix of software engineers and algorithm specialists with backgrounds in optimization, robotics, applied math, and large-scale simulation. We work primarily in C++ and Python, with a strong emphasis on high-quality code, rigorous testing, and reproducible experiments.

As an intern, you will collaborate closely with Dispatch engineers and cross-functional partners (e.g., Motion, Simulation, Data, and Commercial) on problems that directly affect system performance and commercial projects.

Roles & Responsibilities:

In this internship, you will:

• Prototype and evaluate fleet optimization algorithms for problems like vehicle rebalancing, charging strategies, and maintenance/cleaning scheduling (e.g., using mixed-integer optimization, dynamic programming, or heuristic/metaheuristic methods).
• Explore reinforcement learning–based approaches for selected dispatch decisions (e.g., when to send vehicles to charge, how to route vehicles through busy junctions), including state representation, reward design, and basic policy evaluation in simulation.
• Design and run simulation experiments to compare algorithm variants (optimization- or RL-based) using metrics such as wait time distributions, fleet utilization, energy usage, and robustness under disruptions.
• Contribute production-quality code to the Dispatch codebase in C++ and/or Python, including unit tests, integration tests, and clear documentation.
• Collaborate with teammates to translate high-level operational or commercial questions (e.g., “How many vehicles do we need for this project?” or “What if charging is slower?”) into well-posed optimization or simulation studies.
• Work with other autonomy and platform teams to understand constraints coming from motion limits, energy usage, and infrastructure design, and incorporate them into your models and algorithms.

Participate in code reviews and design discussions, giving and receiving feedback to improve both code quality and overall system design.

Knowledge, Skills and Abilities:

We’re looking for someone who brings:

• Academic background in computer science, operations research, robotics, electrical engineering, applied mathematics, or a related field.
• Current undergraduate (rising senior) or graduate student status (MS or PhD) with relevant coursework or research in optimization and/or reinforcement learning.
• Solid programming skills in at least one of:
• C++ (preferred for production code), and/or
• Python (preferred for prototyping, data analysis, and RL/optimization experiments).
• Coursework or experience in optimization, such as:
• Linear / integer / mixed-integer programming
• Dynamic programming, approximate dynamic programming, or stochastic optimization
• Heuristics or metaheuristics (e.g., simulated annealing, genetic algorithms, search-based methods)
• Coursework or experience in reinforcement learning, such as:
• Markov decision processes, value-based and/or policy-based methods
• Function approximation (e.g., neural networks) and experience with a framework like PyTorch or TensorFlow is a plus
• Experience training and evaluating RL policies in simulated environments is a plus
• Strong grasp of algorithms, data structures, and complexity, and comfort reasoning about performance trade-offs in large-scale systems.
• Familiarity with probability, statistics, and simulation, including designing experiments and interpreting results.
• Software engineering fundamentals:
• Comfort working in a Linux environment
• Experience with version control (git) and collaborative development workflows
• Writing clear, maintainable, and tested code
• Ability to communicate technical ideas clearly, both in writing and in discussions, and to collaborate effectively with teammates from different disciplines.

Glydways provides equal employment opportunities to all employees and applicants for employment and prohibits discrimination and harassment of any type without regard to race, color, religion, age, sex, national origin, disability status, genetics, protected veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by federal, state or local laws.