Senior GenAI Full-Stack Engineer

New

CoduranceAI customer support

BrazilFull-TimeSenior

Salary not disclosed

Apply NowOpens the employer's application page

Job Details

Have proven experience building and operating production LLM-powered systems such as chatbots, AI assistants, agent copilots, RAG systems, or LLM orchestration platforms.
Bring strong TypeScript and Node.js engineering skills, including TypeScript strict-mode fluency.
Have production AI experience with prompt engineering, RAG pipelines, agent design, tool calling, model evaluation, observability, and failure-mode analysis.
Be comfortable working across NestJS APIs, React UIs, databases, infrastructure, and production operations.
Be able to evaluate tradeoffs between model quality, latency, reliability, throughput, and cost.
Be able to troubleshoot AI systems across prompts, retrieval pipelines, model configuration, infrastructure, and application code.
Apply state-machine thinking to complex asynchronous workflows; XState or similar experience is a strong signal.
Understand REST API design, asynchronous patterns such as queues and events, and caching strategies.
Have a strong testing culture, including unit, integration, and contract tests.
Have experience working in a monorepo with multiple interconnected services.
Hands-on MCP experience or experience building tool-use agentic workflows is a strong plus.
Experience with LLM observability and evaluation platforms, operating AI workloads at scale, or evaluating multiple foundation models and providers is a strong plus.

Design and extend production LLM applications and agentic workflows using NestJS, XState v5, and the OpenAI SDK.
Build workflows for RAG, intent detection, clarification, fulfillment, escalation, tool use, and human-in-the-loop processes.
Build and maintain the conversation-machine substrate, including guard and action registries, flow validation with ajv, DB-driven flow configurations, and design-time tooling in Epicenter admin.
Build and evolve the Epic Support Assistant and Agent Support Assistant.
Integrate with MCP servers for tool use and agentic behaviors.
Evaluate, benchmark, and tune models across providers including OpenAI, Gemini, and Anthropic.
Troubleshoot production LLM issues such as hallucinations, retrieval failures, prompt regressions, model drift, latency bottlenecks, and provider outages.
Build resilience mechanisms including retries, fallback routing, caching, streaming, rate limiting, and provider routing.
Instrument and tune model quality using Langfuse, evaluation datasets, A/B testing, prompt versioning, and production telemetry.
Manage asynchronous workloads with BullMQ, caching with Redis, and PostgreSQL persistence via Kysely.

View Full Description & ApplyYou'll be redirected to the employer's site