About this role
Design and extend production-grade LLM applications and agentic workflows using
NestJS, XState v5, and the OpenAI SDK — flows include RAG, intent detection,
clarification, fulfillment, escalation, tool-use, and human-in-the-loop state machines
- Build and maintain the conversation-machine substrate: guard/action registries, flow
validation (ajv), DB-driven flow configs, and design-time tooling
- Build and evolve the AI systems behind Support Assistant , the
player-facing support chatbot, and Agent Support Assistant, the AI copilot used by
customer support agents
-
Integrate with MCP servers (Model Context Protocol) for tool-use and agentic behaviors
-
Evaluate, benchmark, and tune models across providers including OpenAI, Gemini,
Anthropic, and future providers; own model selection decisions balancing quality,
latency, throughput, reliability, and cost
- Troubleshoot production LLM issues including hallucinations, retrieval failures, prompt
regressions, model drift, token inefficiencies, latency bottlenecks, and provider outages
- Build resilience mechanisms: retries, fallback routing, caching, streaming, rate limiting,
and provider routing
- Instrument and tune model quality using Langfuse (tracing, evals, prompt
management), evaluation datasets, A/B testing, prompt versioning, and production
telemetry
- Manage async workloads via BullMQ and caching with Redis; PostgreSQL persistence
via Kysely
Requirements
-
Must-Have
-
Proven experience building and operating production LLM-powered systems
-
similar in scope to chatbots, AI assistants, agent copilots, RAG systems, or LLM
-
orchestration platforms
-
Strong TypeScript/Node.js engineering; TypeScript strict-mode fluency
-
Production AI experience: prompt engineering, RAG pipelines, agent design, tool
-
calling, model evaluation, observability, and failure-mode analysis — you've shipped AI
-
features, not just prototyped them
-
Fullstack depth: comfortable moving between NestJS APIs, React UIs, databases,
-
infrastructure, and production operations; you don't artificially limit yourself to one layer
-
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
-
State machine thinking — you naturally model complex async workflows; XState or
-
similar experience is a strong signal
-
Solid understanding of REST API design, async patterns (queues, events), and caching
-
strategies
-
Strong testing culture: unit, integration, and contract tests are first-class deliverables, not
-
afterthoughts
-
Experience working in a monorepo with multiple interconnected services
-
Strong Plus
-
Hands-on experience with MCP (Model Context Protocol) or building tool-use agentic
-
workflows
-
Familiarity with Langfuse or other LLM observability/evaluation platforms
-
Experience operating AI workloads at scale
-
Experience evaluating multiple foundation models and providers
-
Experience building AI copilots, assistants, or conversational products
-
Experience with semantic search and retrieval architectures
-
Experience with AI gateways such as Portkey or similar platforms
-
Experience with NestJS specifically: modules, providers, guards, interceptors, DI
-
patterns
-
Background in customer support or player support platforms — you understand the
-
stakes of getting AI-generated responses wrong
-
Experience shipping under low-latency constraints (chatbot response time budgets,
-
streaming)
-
Previous work in gaming or high-volume consumer products