AI Engineer

PlayStation Global•San Diego, CA

13h•$150,000 - $225,000•Hybrid

About The Position

Sony Interactive Entertainment (SIE) PlayStation is seeking an AI Software Engineer to join the Store Publishing Operations & Commerce (SPOC) team. This role focuses on applied AI engineering, using modern AI systems to solve business problems and create reusable capabilities for safe, reliable, and efficient AI application. The SPOC team ensures trust, security, and seamless transactions across PlayStation's digital store, providing data-driven systems and AI capabilities for platform services, customer experience, payments, fraud prevention, risk management, and business operations. The AI Software Engineer will design, build, and support production AI capabilities that address high-value business problems within SPOC and the broader digital commerce ecosystem. This is an applied AI engineering role focused on turning AI into dependable products, services, and automation, rather than a model-training or predictive-platform ownership role. The engineer will collaborate with various partners to build AI-powered workflows that enhance speed, quality, insight, and decision support, potentially including LLM-powered services, RAG, agentic workflows, tool/function calling, structured outputs, evaluation harnesses, guardrails, and reusable AI platform components. The ideal candidate is a strong software engineer with backend or platform experience and practical applied AI experience, comfortable with foundation model APIs, vector search, prompt evaluation, AI observability, and cloud services, and eager to learn and contribute to reliable production systems.

Requirements

Bachelor's degree in computer science, engineering, a related technical field.
Hands-on experience building AI or generative AI features that connect model APIs to business workflows, data, documents, or internal services.
Strong software engineering skills in Python and/or Java, including API development, testing, debugging, asynchronous processing, and maintainable service design.
Experience with AWS or equivalent cloud services such as EKS/ECS, Lambda, S3, DynamoDB, Kinesis, EventBridge, SNS, SQS, IAM, or related cloud infrastructure.
Experience using model APIs and platforms such as OpenAI/Azure OpenAI, Anthropic Claude, Google Gemini or Vertex AI, AWS Bedrock, Mistral, Meta Llama, or comparable services.
Familiarity with embeddings, chunking, indexing, retrieval strategies, vector and hybrid search, reranking, citations, and vector stores such as OpenSearch, Pinecone, Weaviate, Redis, pgvector, Azure AI Search, or similar technologies.
Experience or familiarity with AI orchestration patterns and tools such as LangChain, LangGraph, LlamaIndex, Semantic Kernel, OpenAI Agents SDK, AWS Bedrock Agents and Knowledge Bases, or comparable tools.
Experience designing prompts, schemas, tool/function calls, workflow contracts, and validation logic so AI systems can produce dependable outputs and interact safely with internal systems.
Familiarity with tracing, monitoring, evals, prompt testing, quality metrics, and debugging tools such as LangSmith, Arize Phoenix, OpenTelemetry, Datadog, Splunk, New Relic, CloudWatch, or comparable platforms.
Understanding of enterprise data security, access control, privacy, responsible AI practices, auditability, and safe handling of sensitive commerce and customer data.
Ability to turn business requirements into technical tasks, collaborate across teams, and explain AI tradeoffs in clear, practical terms.

Nice To Haves

Familiarity with Model Context Protocol (MCP) or similar patterns for connecting AI applications to enterprise tools, databases, documents, and workflows.
Experience contributing to model gateways, prompt registries, reusable tool libraries, workflow templates, evaluation dashboards, or other shared capabilities that help teams adopt AI consistently.
Experience with text, image, document, audio, or video models, including multimodal embeddings, OCR/document understanding, or content moderation workflows.
Exposure to safety alignment, policy enforcement, prompt-injection defenses, data loss prevention, content filtering, red teaming, and human review workflows for AI products.
Familiarity with Kafka, Flink, Spark, dbt, Airflow, or similar data and orchestration tools where they support AI-powered workflows and operational automation.
Experience applying AI to fraud, payments, risk, customer support, marketplace operations, trust and safety, content operations, or digital commerce business processes.

Responsibilities

Build Applied AI Features: Implement services, workflows, and reusable components for LLM-powered automation, retrieval, tool use, summarization, classification, decision support, and knowledge workflows.
Solve Business Problems with AI: Collaborate with operations, product, data, risk, and engineering stakeholders to understand use cases, prototype solutions, measure outcomes, and help move proven capabilities into production.
Support Agentic Workflows and Integrations: Build AI workflows that use tool/function calling, structured outputs, workflow state, internal APIs, and human review patterns to take useful action while staying auditable and controlled.
Develop Retrieval and Knowledge Systems: Contribute to RAG and agentic retrieval pipelines over enterprise content and operational data using embeddings, vector databases, hybrid search, reranking, citations, access controls, and freshness strategies.
Improve AI Quality, Safety, and Evaluation: Create and maintain evaluation suites, regression tests, prompt/model versioning, trace analysis, guardrails, policy checks, PII handling, hallucination mitigation, and operational monitoring.
Production AI Engineering: Develop scalable APIs, microservices, and event-driven workflows in Python or Java, with attention to reliability, resilience, security, cost efficiency, and clean integration with existing services.
Cloud Delivery and Automation: Deploy AI services using AWS, containers, infrastructure as code, CI/CD pipelines, secrets management, observability, and operational runbooks.
Cross-Functional Collaboration: Participate in design reviews, implementation planning, troubleshooting, documentation, and knowledge sharing across technical and non-technical teams.