Job Summary:
We are seeking an AI-Native Software Engineer who views AI not just as an autocomplete tool, but as a core collaborative partner in software delivery. In this role, you will spend less time manually writing boilerplate and more time architecting systems, designing precise technical specifications, and orchestrating multi-agent workflows.
Core Responsibilities
- System Architecture & Design: Define high-level system structures, API contracts, and data models before instructing AI tools to implement them. Own the design, not just the execution.
- Context Engineering & Spec Writing: Author rigorous, unambiguous technical specifications and context rules to guide AI agents toward deterministic, reviewable outputs.
- RAG Pipeline Design: Architect and own end-to-end Retrieval-Augmented Generation pipelines, document ingestion, chunking strategy, embedding selection, vector store configuration, hybrid retrieval, and relevance evaluation.
- Agentic Workflow Management: Build and operate agent harnesses using orchestration frameworks (e.g. LangGraph, LangChain, AutoGen) including tool definitions, routing logic, guardrails, fallback paths, and evaluation hooks.
- Human-in-the-Loop Validation: Design and enforce HITL gates for agentic write operations. Know when to automate and when to require human sign-off, especially for irreversible or high-stakes actions.
- Review, test, and audit AI-generated code for security vulnerabilities, performance characteristics, edge cases, and architectural alignment before it reaches production.
Required Technical Skills
- Engineering Fundamentals: Strong mastery of computer science fundamentals — data structures, algorithms, distributed systems, and system design. You must be able to catch and correct AI errors because you understand the underlying systems.
- Code Review & Auditing: Exceptional ability to read, evaluate, and critique AI-generated code across multiple languages rapidly.
- Agentic System Design: Hands-on production experience building agent harnesses, multi-agent orchestration pipelines, and supervisor/routing patterns using frameworks such as LangGraph, LangChain, or equivalent.
- RAG & Retrieval Engineering: Practical experience designing RAG pipelines including vector store selection, embedding strategies, hybrid search, Reciprocal Rank Fusion, and retrieval quality evaluation.
- AI Tooling Proficiency: Advanced hands-on experience with AI-native IDEs (e.g. Cursor, Windsurf, GitHub Copilot) and command-line agentic tools (e.g. Claude Code, Aider, Codex CLI).
- Context & Prompt Engineering: Proven ability to manage AI context windows, system instructions, tool schemas, and prompt structure to produce consistent, auditable outputs.
- Cloud & API Integration: Solid experience with cloud-native deployment (Azure, AWS, or GCP), RESTful API design, async patterns, and enterprise identity/auth integration.
- Testing & CI/CD: Strong experience writing automated test suites to validate AI-generated logic inside modern CI/CD pipelines, including adversarial and edge-case coverage.
Preferred Qualifications
- Bachelor's or Master's degree in Computer Science, Software Engineering, or equivalent deep production experience.
- Experience integrating with enterprise HR, workforce, or ERP platforms (e.g. SAP SuccessFactors, Workday, Concur, or Oracle HCM) — particularly in an agentic or API integration context.
- Hands-on ML experience beyond API consumption: model fine-tuning, training pipelines, evaluation frameworks, or MLOps deployment.
- Familiarity with enterprise identity providers (e.g. OKTA, Azure AD) and secure token handling in agentic contexts.
- A portfolio or GitHub repository demonstrating projects built primarily via agentic or spec-driven development methodologies.
关于高知特 (Cognizant)
高知特(Cognizant)（纳斯达克代码：CTSH）作为一家AI Builder和相关技术服务提供商，致力于通过打造全栈AI解决方案，帮助企业将人工智能投资转化为实际价值。公司凭借深厚的行业经验、流程优化和工程技术专长，将企业独特的业务场景融入科技系统，赋能组织释放人才潜能，推动切实成果，并帮助全球企业在瞬息万变的环境中保持领先。如需了解更多详情，敬请访问 cognizant.ai 或关注@cognizant。
补充雇佣信息
薪酬信息截至本职位发布之日为准。Cognizant 保留在适用法律允许的范围内随时修改该信息的权利。
申请人可能需要通过现场面试或视频会议的方式参加面试。此外，候选人在每次面试时可能需要出示其当前所在州或政府签发的有效身份证件。
Cognizant 是一家提供平等就业机会的雇主。在招聘过程中，您的申请和候选资格不会因种族、肤色、性别、宗教、信仰、性取向、性别认同、国籍、残疾、遗传信息、怀孕、退伍军人身份或任何其他受联邦、州或地方法律保护的特征而受到影响。
