Senior Software Engineer AI Platform

About the Role

Mirai, a Scopely Studio, is seeking a Senior Software Engineer, AI Platform, to join our team in Riyadh, Riyadh Province. This full-time role focuses on building engaging gaming experiences and developing the underlying AI platform. The ideal candidate will have 5-10 years of experience in software development, with a strong background in creating shared services and libraries from product work.

AI Native Engineering Team Focus

Our AI Native Engineering team develops software solutions for Scopely game teams and internal departments. This includes tools leveraging large language models (LLMs) for bug triage, player feedback analysis, market research, and contract analytics, alongside direct-to-consumer web stores for Scopely games. The core pattern involves connecting source systems, building trusted data stores, applying LLMs or governed logic, and managing access controls. This role is focused on software engineering, not infrastructure, with DevOps and SRE engineers handling clusters, pipelines, and cloud provisioning.

Key Responsibilities

• Build end-to-end product features for internal AI tools using TypeScript or Python, from design to production.
• Identify common patterns across products and transform them into shared services and libraries, including model access, identity, permissions, retrieval, and LLM call controls.
• Develop MCP tool servers to expose company data to agents and chat tools, ensuring user permissions are enforced before data reaches the model.
• Implement relationship-based access control across products for consistent application to database queries and vector search.
• Standardize model access via the LLM gateway to centralize model choice, cost, and usage.
• Extend safety patterns, such as prompt-injection handling, kill switches, sandboxed execution, and locked-down data access, to all products.
• Ensure shared components are easily adoptable through clear APIs, versioning, templates, and comprehensive documentation.
• Set standards for code review, including AI-assisted code, and mentor engineers through review, pairing, and design.
• Manage incidents related to developed components and collaborate with DevOps engineers on deployment and operational aspects.

Qualifications and Experience

• Six or more years of experience building and operating production software, including at least one shared service, library, or internal API used by other engineers or teams.
• Proficiency in TypeScript, ****, and Python.
• Experience shipping LLM features to production, including retrieval over embeddings, tool calling, structured output, and output quality validation.
• Demonstrated ability to design APIs or libraries for other engineers, with attention to versioning, compatibility, safe defaults, and documentation.
• Experience building multi-tenant or permission-sensitive systems, understanding access enforcement to prevent unauthorized data access by users or models. Treats model output and retrieved content as untrusted input.
• Solid understanding of system design principles: service boundaries, data modeling, failure handling, idempotency, and cost.
• Production experience with AWS, including infrastructure as code (Terraform or similar), sufficient to manage service deployment alongside DevOps.
• Practice writing tests as part of development and treating CI checks as mandatory gates.
• Clear written and spoken English, with the ability to explain technical tradeoffs to product and business stakeholders.
• Proficiency in using AI coding tools and verifying their output, taking ownership of merged code.

Technical Environment

The technical stack includes:

• Languages: TypeScript, **, Python
• Services: Express, FastAPI, *****
• AI: OpenAI models and embeddings, LangChain, access control with SpiceDB.
• Internal Tools: Model Context Protocol (MCP) SDK, an internal LLM gateway built on LiteLLM over Amazon Bedrock.
• Cloud and Delivery: AWS (EC2, ECS Fargate, Lambda, App Runner, S3, CloudFront), Terraform, GitLab, GitHub.
• Data: PostgreSQL (Amazon RDS and Aurora), BigQuery, Amazon OpenSearch Serverless for vector search.
• Identity and Access: Okta, OAuth with PKCE, relationship-based access control with SpiceDB.
• Work Tracking: Jira, Confluence.