Agentic AI Engineer
? Remote (LATAM)
Apply:
[email protected] (Subject: App for Agentic AI Engineer)
About Zapp:
Zapp is transforming how immigrants send money home, making transfers fast, fair and simple, directly through WhatsApp. We’re a high-growth fintech serving Hispanic families across the US.
About the Role:
We’re looking for an Agentic AI Engineer who goes beyond prompting and builds the AI infrastructure that runs the company. You’ll architect autonomous agents, connect LLMs to our APIs, data and business workflows, and deploy intelligent automation across Product, Ops, Compliance, Risk and Data.
Production matters here. You’ll take agents from experiment to reliable, secure and cost-efficient systems that real operations depend on.
What you’ll do
- Agentic architecture: design and ship multi-step agent workflows that reason, plan and interact with internal tools and external APIs, using LangGraph, OpenAI Assistants, LlamaIndex, n8n, Make or similar.
- Multi-model strategy: integrate and evaluate LLM providers (Anthropic, OpenAI, Gemini, Qwen, etc.) with dynamic routing, fallbacks and evaluation pipelines that balance accuracy, latency and cost.
- MCP development: build and secure MCP servers that expose internal capabilities to agents, such as database access, file retrieval, validation utilities and business-logic tools.
- Prompt engineering: craft structured prompts, tool schemas and evaluation datasets, and maintain versioned prompt libraries with quality benchmarks.
- RAG and integrations: connect agents to backend services, vector stores and retrieval pipelines with the Engineering and Data teams.
- Observability and safety: set up monitoring, logging, guardrails and alerting so agents behave predictably and meet regulatory requirements.
- Optimization: debug and profile agents to reduce latency, cost and token usage.
- Documentation:
keep clear standards for agent workflows, MCP tools, prompts and integration patterns.
What we’re looking for
- Fully bilingual in English and Spanish.
- 7+ years in software or automation engineering, including 2+ years working directly with LLMs, agent frameworks or AI-powered automation.
- Proven production integrations using LLMs, APIs and workflow tools, not just prototypes.
- Strong proficiency in Python and/or TypeScript/Node.js.
- Deep familiarity with LangChain, LangGraph, LlamaIndex, OpenAI Assistants or similar ecosystems.
- Hands-on experience with n8n, Make or Zapier.
- Solid understanding of vector databases (Pinecone, Qdrant, Redis), embeddings and RAG.
- Experience building or securing MCP tools for data access, business logic or internal APIs.
- Knowledge of API design, backend services, cloud environments and CI/CD.
- A builder’s mindset. You’re comfortable with ambiguity and creating systems from scratch.
- Clear communication. You turn ambiguous business needs into concrete technical solutions.
Nice to have
- Experience in fintech, digital payments or other regulated environments.
- Multi-model architectures with routing and evals in production.
- Observability for AI pipelines (Langfuse, LangSmith, Datadog, Grafana or similar).
- Experience automating operational processes like compliance, risk or customer support.
Why Zapp
- Real ownership from day one and fast decisions.
- A lean, senior team across Product, Engineering, Analytics and Growth.
- The chance to build the company’s agentic infrastructure from the ground up.
- AI at the core of how the whole company operates, not a side project.
- Work that directly helps immigrant families.
- Remote-first across LATAM.
#Hiring #AIEngineer #AgenticAI #LLM #MCP #LangGraph #LangChain #RAG #GenerativeAI #n8n #Python #Fintech #RemoteJobs #LATAM #Bilingual #Empleo #TrabajoRemoto
📌 Agentic AI Engineer (Colombia)
🏢 Zapp
📍 Colombia