Software Engineer, Scanless Tech - Applied Automation

About the role

Scanless Tech is building the future of item identification in Amazon's vast fulfillment network. We're eliminating the need for explicit, manual scans by teaching robotic systems to see and understand the physical world autonomously. Our systems operate at massive scale across dozens of robotic platforms, processing millions of identification events daily in environments where speed, accuracy, and reliability are all non-negotiable. This role provides ample opportunity to influence, innovate and shape the next generation of robotics tech. You will partner with other software and hardware teams at Amazon while being on the forefront of new product development and technology that improves our systems and the experience for our customers around the globe.

This role sits at the intersection of physical systems and artificial intelligence. You'll build software that gives robotic workcells the ability to perceive, interpret, and act on visual information in real time, without human intervention. Our systems must reason about objects in unstructured, dynamic environments where lighting changes, orientations vary, and milliseconds matter. You'll work with sensor hardware, edge compute, and cloud infrastructure to deliver intelligence that operates reliably at industrial scale.

Key job responsibilities

• Design, build, and maintain production ML/CV inference pipelines that run on edge hardware inside fulfillment workcells
• Develop and optimize model training pipelines, including data collection, labeling, augmentation, and evaluation
• Own end-to-end delivery of features from design through deployment, monitoring, and operational excellence
• Collaborate with engaged, friendly, fundamentals-first thinking oriented engineers and applied scientists to translate research prototypes into production-grade systems for customer use cases and implement core engineering capabilities for Amazon's enormous fulfillment network
• Integrate perception systems with robotic platforms and sensor hardware (cameras, lighting, compute modules)
• Define and track performance metrics; investigate and resolve production issues across a fleet of deployed systems
• Contribute to system architecture decisions that balance accuracy, latency, cost, and maintainability
• Drive operational improvements by operating across multiple layers of the stack to design and build fault-tolerant production software that optimize our fleet of fulfillment infrastructure and robots
• Hire and mentor other engineers on the team

A day in the life

We are developing state-of-the-art technologies that will forever change the way products and packages are identified, all by using Computer Vision and Machine Learning. We envision a time when manual item identification is a vestige of the past. As a Software Development Engineer (SDE) on this exciting and diverse team, you will develop and deploy ML and CV models that solve real-world perception problems in Amazon's fulfillment network. This includes training and optimizing deep learning models for visual recognition tasks, building data pipelines that continuously improve model performance, and designing inference systems that meet strict latency and accuracy requirements on constrained edge devices. You'll collaborate closely with applied scientists to move research into production and measure impact through real-world metrics.

On any given day, you might be profiling a new model architecture on edge hardware, balancing accuracy gains against latency budgets. The next moment, you could be investigating a fleet-wide performance anomaly using real production data, or designing a new augmentation strategy informed by failure cases you discovered in the field. You'll review architecture proposals with partner robotics teams onboarding your system, debate trade-offs in a code review with teammates, and ship changes that impact millions of identification events across Amazon's network. The problems are tangible, you can walk into a fulfillment center and watch your code run and no two days look the same.

Basic qualifications

• 3+ years of non-internship professional software development experience
• 2+ years of non-internship design or architecture (design patterns, reliability and scaling) of new and existing systems experience
• 1+ years of software development engineer or related occupational experience
• 1+ years of designing and developing large-scale, multi-tiered, multi-threaded, embedded or distributed software applications, tools, systems, and services using: C#, C++, Java, or Perl experience
• 1+ years of Object Oriented Design experience
• Bachelor's degree or foreign equivalent in Computer Science, Engineering, Mathematics, or a related field
• Experience programming with at least one software programming language

Preferred qualifications

• 3+ years of full software development life cycle, including coding standards, code reviews, source control management, build processes, testing, and operations experience
• Bachelor's degree in computer science or equivalent
• Knowledge of machine learning model architecture and inference
• Experience with one of the following areas: machine learning technologies, Reinforcement Learning, Deep Learning, Computer Vision, Natural Language Processing (NLP) or related applications
• Experience that includes strong analytical skills, attention to detail, and effective communication abilities
• Decisive with limited information and able to learn rapidly

Pay

USA, MA, Westboro - 143,700.00 - 194,400.00 USD annually

Benefits

Amazon offers a full range of benefits for you and eligible family members, including domestic partners. Benefits can vary by location, the number of regularly scheduled hours you work, length of employment, and job status such as seasonal or temporary employment. The benefits that generally apply to regular, full-time employees include:

• Medical, Dental, and Vision Coverage
• Maternity and Parental Leave Options
• Paid Time Off (PTO)
• 401(k) Plan