Senior GenAI Full-Stack Engineer

Codurance; Brasil; On-site; BRL 200,000 - 350,000; Full time

Job summary

Casado, a Brazil-based AI tech company, seeks an experienced LLM systems engineer to design and operate production-grade AI applications using NestJS, XState v5, and the OpenAI SDK. You will build RAG pipelines, intent detection and human-in-the-loop state machines, and integrate tool-use into agent copilots for customer support contexts.

Join Casado to architect scalable AI systems, tune models across providers, and ensure low-latency, observable workflows.

Qualifications

• Proven experience building and operating production LLM-powered systems similar to chatbots or AI assistants.
• Strong TypeScript/Node.js engineering, with TypeScript strict-mode fluency.
• Production AI experience: prompt engineering, RAG pipelines, toolcalling, observability, and failure-mode analysis.

Responsibilities

• Design and extend production-grade LLM applications and agentive workflows.
• Build and maintain the conversation-machine substrate, guard/action registries, and flow validation.
• Evaluate, benchmark, and tune models across providers; own model selection decisions balancing quality and cost.
• Troubleshoot production LLM issues, and build resilience: retries, caching, and rate limiting.
• Manage async workloads via BullMQ and caching with Redis; PostgreSQL persistence via Kysely.

Skills

TypeScript

Node.js

NestJS

XState

AI systems

RAG

observability

scalability

monorepo

tool use

Tools

OpenAI SDK

Langfuse

BullMQ

Redis

PostgreSQL

Kysely

Job description

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 machinesBuild and maintain the conversation-machine substrate: guard/action registries, flowvalidation (ajv), DB-driven flow configs, and design-time tooling in Epicenter adminBuild and evolve the AI systems behind Epic Support Assistant (ESA), the player-facing support chatbot, and Agent Support Assistant, the AI copilot used by customer support agentsIntegrate with MCP servers (Model Context Protocol) for tool-use and agentic behaviorsEvaluate, benchmark, and tune models across providers including OpenAI, Gemini, Anthropic, and future providers; own model selection decisions balancing quality, latency, throughput, reliability, and costTroubleshoot production LLM issues including hallucinations, retrieval failures, prompt regressions, model drift, token inefficiencies, latency bottlenecks, and provider outagesBuild resilience mechanisms: retries, fallback routing, caching, streaming, rate limiting, and provider routingInstrument and tune model quality using Langfuse (tracing, evals, prompt management), evaluation datasets, A/B testing, prompt versioning, and production telemetryManage async workloads via BullMQ and caching with Redis; PostgreSQL persistence via Kysely

Requirements

• Must-HaveProven experience building and operating production LLM-powered systemssimilar in scope to chatbots, AI assistants, agent copilots, RAG systems, or LLMorchestration platforms
• Strong TypeScript/Node.js engineering; TypeScript strict-mode fluency
• Production AI experience: prompt engineering, RAG pipelines, agent design, toolcalling, model evaluation, observability, and failure-mode analysis — you've shipped AIfeatures, 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, modelconfiguration, infrastructure, and application code
• State machine thinking — you naturally model complex async workflows; XState orsimilar experience is a strong signal
• Solid understanding of REST API design, async patterns (queues, events), and cachingstrategies
• Strong testing culture: unit, integration, and contract tests are first-class deliverables, notafterthoughts
• Experience working in a monorepo with multiple interconnected services
• Strong PlusHands-on experience with MCP (Model Context Protocol) or building tool-use agenticworkflows
• 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, DIpatterns
• Background in customer support or player support platforms — you understand thestakes 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