Senior Gen AI Engineer (Games), Bulgaria

Stack

Responsibilities

• ▹Design and extend production-grade LLM applications and agentic workflows, including RAG, intent detection, clarification, fulfillment, escalation and human-in-the-loop flows
• ▹Build and maintain the conversation engine, including flow validation, configuration tooling and the admin experience
• ▹Integrate tool-use and agentic protocols to extend what the AI assistant and copilot can do
• ▹Evaluate, benchmark and tune models across multiple providers, owning model selection decisions that balance quality, latency, throughput, reliability and cost
• ▹Troubleshoot production AI issues such as hallucinations, retrieval failures, prompt regressions, model drift and provider outages
• ▹Build resilience mechanisms including retries, fallback routing, caching, streaming and rate limiting
• ▹Instrument and tune model quality using tracing, evaluation datasets, A/B testing and prompt versioning
• ▹Build player-facing and support-agent-facing experiences in React, and develop the admin interface used to manage flows, intents, releases, translations, prompts and evaluations
• ▹Integrate real-time, event-driven product features and own frontend performance, accessibility and internationalization
• ▹Write comprehensive automated tests across unit, integration, contract and end-to-end layers
• ▹Build observability for traditional services and AI systems, including tracing, dashboards and alerting
• ▹Diagnose performance bottlenecks across application code, retrieval systems, model inference and infrastructure
• ▹Participate in an on-call rotation and respond to production incidents with clear communication
• ▹Document architectural decisions and keep runbooks current

Requirements

• ▹Proven experience building and operating production LLM-powered systems: chatbots, AI assistants, agent copilots, RAG systems or LLM orchestration platforms
• ▹Strong TypeScript/Node.js engineering skills, with fluency in strict-mode TypeScript
• ▹Hands-on production AI experience: prompt engineering, RAG pipelines, agent design, tool calling, model evaluation, observability and failure-mode analysis
• ▹Full-stack depth: backend APIs, React UIs, databases, infrastructure and production operations
• ▹Ability to evaluate tradeoffs between model quality, latency, reliability, throughput and cost
• ▹Ability to troubleshoot AI systems across prompts, retrieval pipelines, model configuration, infrastructure and application code
• ▹Experience modeling complex asynchronous workflows (state machine experience is a strong signal)
• ▹Solid understanding of REST API design, asynchronous patterns and caching strategies
• ▹A strong testing culture, with unit, integration and contract tests treated as first-class deliverables
• ▹Experience working in a monorepo with multiple interconnected services

Nice to have

• ▹Experience with agentic tool-use protocols (e.g. MCP), semantic search/retrieval architectures or AI gateways
• ▹Familiarity with LLM observability and evaluation platforms, and experience evaluating multiple foundation models or providers at scale
• ▹Experience building AI copilots, assistants or other conversational products, especially under low-latency or streaming constraints
• ▹Hands-on experience with NestJS
• ▹Background in customer support or player support platforms, gaming or high-volume consumer products

Soft skills

About the company

TELUS Digital is the digital division of Canada's TELUS, an award-winning digital product consultancy offering web, mobile, enterprise AI and customer care technology services with a global team.