Machine Learning Developer

MIT Lincoln Laboratory•Lexington, MA

8d•$145,200 - $220,000•Hybrid

About The Position

The Transportation Safety and Resilience Group develops integrated sensing and decision support systems that enable a safe and resilient global transportation system. We specialize in systems analysis, prototyping, system architectures, and algorithm development to support supply chain resilience, improving safety in the current airspace and the introduction of new and more autonomous vehicles into the existing transportation system. A current focus area of our work is the safe and efficient introduction of Advanced Air Mobility vehicles and other new entrants into the National Airspace System. The systems we work on help prevent aircraft accidents in the existing airspace and are extensible to a more autonomous future airspace. We are also focused on improving the resilience and effectiveness of the Department of War supply chain. With technical expertise in real-time software architecture, systems analysis, advanced supercomputing-enabled modeling and simulation, machine learning, and system integration, our research teams take new ideas for solving problems and develop them into working prototypes. We are currently working on detect-and-avoid systems for both crewed and uncrewed aircraft systems, safety systems to prevent runway incursions, and decision support systems. We are looking for applicants with an interest and background in applied engineering, modeling and simulation and machine learning to develop algorithms and systems for autonomous and semi-autonomous vehicles. Additionally, we are looking for applicants to develop new advanced decision support systems as well as architectures and techniques for advancing and deploying semi-autonomous systems in the broader transportation domain. Example problem areas include algorithm development for collision avoidance logic, real-time contingency management, inter-vehicle coordination, and validation of decision support, surveillance, and tracking systems. Successful candidates will develop the skills to analyze the operational problems in detail and to develop deployable solutions, while extending our impact to other areas of the transportation domain. The candidate should be familiar with developing solutions using modern machine learning approaches such as reinforcement learning. Additional responsibilities include leading and developing applied research and development programs in the transportation domain, formulating new approaches to existing challenges and publishing results in conferences/journals. The Transportation Safety and Resilience Group supports a hybrid work environment.

Requirements

PhD in engineering, physics, Mathematics, or computer science or similar field; in lieu of a PhD, a Master’s degree with four years of relevant experience will be considered.
Experience with state of the art machine learning approaches and architectures
Experience developing algorithms and architectures for advancement and deployment of autonomous systems
Experience with reinforcement learning
Publication record with conference and/or journal articles
Experience with algorithmic software development in Python
Ability to collaborate well as part of a team with good interpersonal skills
Ability to effectively distill and distribute concepts and results to a wide audience

Nice To Haves

Knowledge of the air or surface transportation domain
Experience developing algorithms and/or mathematical models in a simulation environment and/or deploying and testing on hardware
Experience with Natural Language Processing, Natural Language Understanding or Automatic Speech Recognition
Experience in other programming languages such as Python, C++, MATLAB, Julia, etc.
Publications in (air) transportation related conferences/journals
Experience leading a small team
Experience with distributed computing and scalable machine learning frameworks
Experience submitting research proposals

Responsibilities

Develop algorithms and systems for autonomous and semi-autonomous vehicles.
Develop new advanced decision support systems.
Develop architectures and techniques for advancing and deploying semi-autonomous systems in the broader transportation domain.
Develop algorithms for collision avoidance logic, real-time contingency management, inter-vehicle coordination, and validation of decision support, surveillance, and tracking systems.
Analyze operational problems in detail and develop deployable solutions.
Extend impact to other areas of the transportation domain.
Develop solutions using modern machine learning approaches such as reinforcement learning.
Lead and develop applied research and development programs in the transportation domain.
Formulate new approaches to existing challenges.
Publish results in conferences/journals.