Software Development Engineer II, Device OS

We need to output HTML fragment with sections for headings: About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. Only include sections where posting has real specific content. We must not repeat job title. No Job ID etc. Remove boilerplate. We need to parse the description. Key sections: - About the role: The description starts with "Description The Amazon Devices Software & Services team is looking for a Software Development Engineer...". That's about the role. Could be "About the role". We'll put that as

About the role

then

with prose. - Responsibilities: There is a "Key job responsibilities" repeated. We'll extract bullet points. The responsibilities list includes many sentences. We'll put them in

• . - Requirements: There's "Basic Qualifications" and "Preferred Qualifications". Those are qualifications. Could be "Requirements" and "Qualifications"? The instruction says use headings: About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. So we need to decide which content goes where. We have Basic Qualifications (required) and Preferred Qualifications (preferred). Could map Basic Qualifications to "Requirements" and Preferred Qualifications to "Qualifications". Or maybe "Skills" for specific technical skills. Let's think. Basic Qualifications list: - 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 Also there is mention of "Experience programming with at least one software programming language" (maybe skill). Could put under Skills. We also have mention of "AI coding assistants" etc but that's more responsibilities. We need to include Benefits and Pay if present. The posting includes: "The base salary range for this position is listed below. Your Amazon package will include sign-on payments and restricted stock units (RSUs). Final compensation will be determined based on factors including experience, qualifications, and location. Amazon also offers comprehensive benefits including health insurance (medical, dental, vision, prescription, Basic Life & AD&D insurance and option for Supplemental life plans, EAP, Mental Health Support, Medical Advice Line, Flexible Spending Accounts, Adoption and Surrogacy Reimbursement coverage), 401(k) matching, paid time off, and parental leave. Learn more about our benefits at https://amazon.jobs/en/benefits. USA, CA, Irvine - 143,700.00 - 194,400.00 USD annually" Thus we have Pay: salary range. Benefits: list. So we should include Pay and Benefits sections. Schedule: Not mentioned. So omit. Skills: Could include specific languages: C#, C++, Java, Perl, maybe also mention AI coding assistants? But better to list technical skills: C#, C++, Java, Perl, OOP, distributed systems, etc. We'll extract from qualifications. We must not invent or summarize away detail. Must preserve ALL substantive information. So we need to include all details from the posting, but we can format them. We must not include boilerplate like EEO, equal opportunity, scam warning, navigation, cookie, legal, duplicated lines, application instructions. Also remove the Los Angeles County specifics? That's part of the posting but maybe considered boilerplate? It's location-specific legal text. Might be considered boilerplate? The instruction: Remove cookie/navigation/legal/EEO/equal-opportunity/scam-warning boilerplate, duplicated lines, and application instructions. The Los Angeles County text is legal; likely should be removed. Also the equal opportunity employer statement is boilerplate; remove. Also the accommodation link etc. Remove. Also remove "If the country/region you’re applying in isn’t listed, please contact your Recruiting Partner." etc. Also remove "Company - Amazon.com Services LLC Job ID: A10560873". Remove Job ID. Also remove "USA, CA, Irvine - 143,700.00 - 194,400.00 USD annually" maybe part of Pay; we keep that as pay. We need to preserve all substantive information: responsibilities, qualifications, benefits, pay, etc. Let's extract responsibilities text: "Key job responsibilities" repeated; we need to capture the bullet-like sentences: - Invent New Paradigms: Move beyond traditional voice UI to build "Ambient Intelligence" experiences that span Alexa, Fire TV, and mobile/wearable surfaces, focusing on proactive assistance rather than reactive commands. - Rapid Innovation: Operate in a fast-paced environment where you will prototype, test, and deploy features that leverage multimodal inputs (camera, audio, and documents) to create personalized customer magic. - Cross-Surface Continuity: Ensure that a customer’s experience is persistent and context-aware beyond the device boundaries - whether they are at their desk using Alexa.com, in their living room with Fire TV, or on the go with a wearable. - Design, build, test, and operate scalable backend services and libraries that power AI/LLM-driven features. - Integrate and harden model providers, tools, and agent capabilities behind clean, versioned interfaces — with sane failure behavior, observability, and cost/latency awareness. - Build evaluation and regression tooling so AI behavior is measured, gated, and trustworthy — not vibes. - Apply security-conscious, fail-closed design to anything touching credentials, user data, tool execution, or autonomous actions. - Work fluently alongside AI coding assistants: decompose tasks an agent can execute, critically review AI-generated code, and keep engineering rigor high at AI speed. Also there is "A day in the life" section: "In our team, you will get to build new products and features from the ground up. You will define and build features and tools used by thousands of developers for building for millions of customers. We work across multiple internal teams as well as product to define what we are building and make it a reality. SDEs on this team are actively writing code, as well as working on designs and mentoring others. As the Device OS org we get to work horizontally across many exciting new device types." And "About The Team": "We’re a platform team that builds the shared AI/agent infrastructure used across multiple products. We care about correctness, safety, and developer experience in equal measure. We operate test-first, review adversarially, keep our core clean and reusable, and treat AI as a tool that raises the engineering bar." We could incorporate these into Responsibilities or maybe a separate "About the role" includes description and team. The instruction: Use

  for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). So we can have About the role (includes description and team). Responsibilities separate. Requirements (basic qualifications). Qualifications (preferred). Skills (maybe languages). Benefits. Pay. We must not create sections with filler. Ensure each section has real content. Let's craft:

  About the role

  The Amazon Devices Software & Services team is looking for a Software Development Engineer to join a high-velocity, specialized team building a new groundbreaking customer experience that will define how millions of users worldwide interact with their technology — making it more proactive, intuitive, and invisible. The team builds the systems that put AI into production — the runtime, services, and tooling that turn large language models and autonomous agents into reliable, safe, scalable products. As a Software Development Engineer II, you’ll design and ship core software in this stack: model and tool integrations, the agent execution loop, evaluation and quality pipelines, and the APIs that product teams build on. You’ll own meaningful components end to end, make real design decisions, and write production code that runs against real users and real data. This is a hands‑on building role for an engineer who wants to go deep on AI infrastructure — not just call an API, but engineer the reliability, safety, and performance around it. You’ll also work in an AI‑native development environment: we use AI coding assistants and agentic workflows as part of our daily software‑development lifecycle, and we hold that velocity to a hard bar of automated testing and code review. We expect engineers who can move fast and keep the quality gates green.

  About the team: We’re a platform team that builds the shared AI/agent infrastructure used across multiple products. We care about correctness, safety, and developer experience in equal measure. We operate test‑first, review adversarially, keep our core clean and reusable, and treat AI as a tool that raises the engineering bar.

  A day in the life: In our team, you will get to build new products and features from the ground up. You will define and build features and tools used by thousands of developers for building for millions of customers. We work across multiple internal teams as well as product to define what we are building and make it a reality. SDEs on this team are actively writing code, as well as working on designs and mentoring others. As the Device OS org we get to work horizontally across many exciting new device types.

  That seems to capture all substantive info about role, team, day. Now Responsibilities:

  Responsibilities

  • Invent New Paradigms: Move beyond traditional voice UI to build "Ambient Intelligence" experiences that span Alexa, Fire TV, and mobile/wearable surfaces, focusing on proactive assistance rather than reactive commands.
  • Rapid Innovation: Operate in a fast‑paced environment where you will prototype, test, and deploy features that leverage multimodal inputs (camera, audio, and documents) to create personalized customer magic.
  • Cross‑Surface Continuity: Ensure that a customer’s experience is persistent and context‑aware beyond the device boundaries — whether they are at their desk using Alexa.com, in their living room with Fire TV, or on the go with a wearable.
  • Design, build, test, and operate scalable backend services and libraries that power AI/LLM‑driven features.
  • Integrate and harden model providers, tools, and agent capabilities behind clean, versioned interfaces — with sane failure behavior, observability, and cost/latency awareness.
  • Build evaluation and regression tooling so AI behavior is measured, gated, and trustworthy — not vibes.
  • Apply security‑conscious, fail‑closed design to anything touching credentials, user data, tool execution, or autonomous actions.
  • Work fluently alongside AI coding assistants: decompose tasks an agent can execute, critically review AI‑generated code, and keep engineering rigor high at AI speed.
  Now Requirements (basic qualifications):

  Requirements

  • 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
  • 1+ years of Object Oriented Design experience
  • Bachelor's degree or foreign equivalent in Computer Science, Engineering, Mathematics, or a related field