Software Development Engineer, Amazon Fulfillment Technology (AFT), Platform Engineering and Services

Amazon Fulfillment Technologies (AFT) powers the physical engine at the heart of Amazon — magically turning your click of a button into packages at your door. Within AFT, the Labor Insights team builds and operates the platform that tracks on associate labor performance across Amazon's entire global warehouse network. Labor Insights' suite of 18+ services supports ~1 million associates and 70,000+ daily portal users across 9,800+ facilities and 10 business lines — ingesting over 19 billion labor events annually to produce the real-time productivity reports that warehouse managers rely on to run their operations every day. We leverage AWS technologies, stream processing, and AI/ML to build distributed, highly available systems that operate at massive scale.

Are you ready for a challenge? As a Software Development Engineer on the Labor Insights team, you will build the next generation of Amazon's labor management platform — designing systems that ingest billions of labor events from warehouse floor tools and time clock systems, aggregate them through multi-stage pipelines, and serve real-time productivity reports to operations managers across the globe. You will own design and development of distributed systems that refresh reports every 10–15 minutes across 3 global regions while maintaining 99.9% availability. You will contribute to a wide range of projects — from event processing pipelines and aggregation engines to reporting APIs, modern React-based portals, and AI-powered configuration tools. Most excitingly, you will help drive refresh of the existing tool, an AI-first greenfield initiative to rebuild the platform from the ground up using modern stream-based architecture, eliminating a decade of technical debt and delivering near real-time metrics. You will be involved in all aspects of the software development cycle including scoping, design, implementation, testing, and delivery. A commitment to teamwork, hustle, and strong communication skills are absolute requirements. Creating reliable, scalable, and high-performance services requires exceptional technical expertise. If you have that expertise, are customer-centric, and motivated about building systems that operate at global scale, then we want to talk to you!

Key job responsibilities

• Design, develop, and test software components across Labor Insights' ingestion, aggregation, reporting, and configuration layers — handling labor events, time clock punches, employee location tracking, and productivity report generation
• Build new services using AI-assisted development practices, leveraging tools like Amazon Q and Bedrock to achieve high-velocity delivery on greenfield systems
• Design large-scale data processing solutions using DynamoDB, PostgreSQL, S3, SQS/SNS, EventBridge, Lambda, and stream processing frameworks (Cosmos/Flink) optimized for high throughput and low-latency access
• Drive service consolidation and modernization — migrating from legacy infrastructure to Native AWS (NAWS) with full CI/CD pipelines
• Participate in definition of coding standards and apply them in code and design reviews
• Provide estimates for development tasks and negotiate timelines with project leaders
• Mentor junior team members and help them acquire critical skills and best practices
• Document and communicate technical concepts to team members and stakeholders
• Participate in deep technical discussions, where you stand up for your ideas and work with team members to find the best solution for the customer
• Identify technical and project risks and take initiative to mitigate them

A day in the life
Full-cycle ownership. You contribute to all aspects of the software development cycle — from scoping and design through implementation, testing, and delivery of features that provide labor performance visibility to Amazon's global fulfillment operations.

Billions-scale data processing. You build systems that ingest 19+ billion labor events per year from warehouse floor tools, time clock systems, and employee location heartbeats. You design multi-stage aggregation pipelines that process this data into PPR and PPA reports refreshed every 10–15 minutes, served to 70K+ daily users across 9,800+ facilities.

AWS-native architecture. You leverage the full AWS stack — DynamoDB for hot data storage, S3 for historical data, PostgreSQL for aggregated reports, SQS/SNS for event-driven processing, EventBridge and Lambda for serverless job orchestration, and Cosmos/Flink for stream processing. You architect for 99.9% availability across 3 global regions (NA, EU, FE).
AI-first development. You pioneer AI-assisted engineering practices — using Amazon Bedrock, Amazon Q, and AI-powered tools to accelerate greenfield development at 20X traditional velocity. You build intelligent systems like AI-powered configuration assistants, automated intake routing, and GenAI-driven operational tooling.

Modernization at scale. You modernize a 15+ year-old platform serving every fulfillment center worldwide — consolidating 23+ services into a clean microservices architecture, rebuilding the portal experience in React from legacy JSP/jQuery, and transitioning from batch to stream processing, all while maintaining zero-downtime operations.

About the team
About Labor Insights: We build Amazon's global labor management platform — the system of record for measuring associate performance across every fulfillment center, sort center, delivery station, and grocery facility worldwide. Labor Insights tracks how ~1 million associates spend their time, producing the productivity reports (PPR, PPA, attendance, Time Off Task) that operations managers use to make staffing decisions, identify bottlenecks, and ensure customer shipments go out on time.

What Makes This Role Exciting:

• Build FCLM 2.0 from the ground up — a once-in-a-career opportunity to greenfield a platform that runs Amazon's global workforce visibility, using AI-first development to deliver in months what would take years through traditional approaches
• Pioneer stream processing — transition from batch to real-time with sub-minute latencies using Cosmos/Flink, fundamentally changing how Amazon tracks labor performance
• Build a multi-tenant platform (FCLM FLEX) — design a configurable translation layer that recasts PPR and PPA as flexible views on the same underlying data, eliminating duplicate pipelines and enabling third-party teams to create custom labor views without engineering support
• Reimagine the portal experience — rebuild the portal used by 70K+ daily users from legacy JSP/jQuery into a modern React application with accessibility, localization, and scalable access controls
• Expand to new businesses — onboard Amazon Grocery (Fresh, Whole Foods) and other new business lines in 2026+, scaling the platform to new node types and labor models
• Build AI-powered tools — design intelligent systems using Amazon Bedrock for automated configuration management, request intake, and operational copilots that serve 1.5M workers globally
• Massive impact — every feature you build is used by tens of thousands of warehouse managers the same day; Labor Insights’ operations save an estimated $25M+ annually

Our tenets:

• Build for Scale: Design generically to support multiple use cases beyond Labor Insights
• Service Generalization for Multi-Tenancy: Build abstractions that work across all businesses
• No Data Filtering by Default: Accept all incoming data, build flexible views on top