AI Engineer Remote Latam – ID #00373

Role Mission

We are seeking a hands-on Senior AI Engineer to design, build, and deploy production-grade AI solutions—including conversational agents, web scraping pipelines, and vector-based retrieval architectures.

This is a hands-on, product-focused engineering role centered on shipping reliable, scalable AI systems in production—not pure research or academic experimentation. You will work closely with backend and frontend teams to transform complex business problems into high-performance, AI-driven applications.

Technical Stack & Architecture Context

• Core Stack: Node.js (JavaScript) with Express backend, React frontend. All internal agents and systems are built within this Node + React ecosystem.
• AI Orchestration & Providers: We do not rely on third-party orchestration frameworks like LangChain or LlamaIndex. Instead, we maintain our own custom orchestration layer. We utilize the OpenAI SDK as a client and route requests across multiple LLM providers (Gemini, GPT-4, Claude, Grok) via OpenRouter based on cost, latency, and quality trade-offs.
• Voice & Realtime: OpenAI Realtime API for interactive, real-time voice agents.
• Data & Search: MongoDB, PostgreSQL (with pgvector), and dedicated vector stores.
• Infrastructure & Cloud: AWS (Lambda, SQS, DynamoDB), Docker, Kubernetes, and event-streaming tools like Kafka.

While tools and frameworks evolve, the single most important requirement for this role is demonstrated, practical experience building and shipping user-facing products with LLMs in production.

Key Responsibilities

• Conversational AI & Agentic Systems: Design and build autonomous AI agents utilizing tool/function calling, structured outputs, strict schemas, and tailored prompt engineering.
• Realtime Voice Systems: Implement interactive speech-to-speech and text-to-speech agentic workflows using the OpenAI Realtime API.
• Retrieval & Semantic Search: Design, optimize, and maintain RAG (Retrieval-Augmented Generation) pipelines using vector databases and embedding models.
• Multi-Model Optimization: Evaluate, route, and benchmark dynamic LLM choices (GPT, Claude, Gemini, Grok) to balance execution cost, response latency, and generation quality per task.
• Data Extraction & Scraping: Develop robust, automated web scraping bots and ETL pipelines for structured and unstructured data collection.
• System Integration & API Design: Expose AI capabilities via secure, high-throughput REST APIs and microservices integrated directly into our core Node.js backend.
• Performance & Reliability: Monitor LLM drift, implement fallbacks, optimize token usage, and ensure uptime for critical workflows.

Required Qualifications & Skills

1. Core Engineering & Backend

• Strong Node.js (JavaScript) Expertise: Proven track record building enterprise backend services using Node.js and Express.
• Custom AI Architectures: Ability to design agentic loops, state machines, and routing logic natively without relying on off-the-shelf frameworks (LangChain/LlamaIndex).
• Databases: Hands-on experience with MongoDB and relational databases like PostgreSQL.
• API Development: Experience designing, building, and scaling microservices and RESTful APIs.

2. Generative AI & Vector Search

• Production LLM Experience: Deep practical expertise with OpenAI, Anthropic, and multi-model routing (OpenRouter).
• Agent Capabilities: Mastery of tool calling / function calling, JSON-structured outputs, dynamic system prompting, and context window management.
• Voice AI: Hands-on experience or working familiarity with Realtime Voice APIs (Speech-to-Text, Text-to-Speech, duplex voice streams).
• Vector Databases & RAG: Practical experience implementing semantic search using vector extensions (pgvector, MongoDB Atlas Vector Search, or dedicated vector stores).
• Practical AI Foundations: Deep understanding of embeddings, tokenization, context compression, cost management, and practical LLM failure modes in production.

3. Scraping & Automation

• Experience building resilient, scalable web scraping bots capable of handling dynamic rendering, rate limits, and proxy management.

Nice-to-Have / Plus Skills

• Cloud Infrastructure (AWS): Hands-on experience with AWS Lambda, SQS, DynamoDB, or S3.
• DevOps & Containers: Experience with Docker containerization, Kubernetes orchestration, and CI/CD pipelines.
• Event-Driven Architectures: Exposure to message brokers and streaming platforms like Apache Kafka.
• Frontend Knowledge: Basic understanding or familiarity with React to effectively collaborate across the full stack.

Soft Skills & Execution Mindset

• Business-to-Tech Translation: Ability to convert vague business objectives into well-defined, robust AI system architectures.
• Pragmatic Problem Solver: Focused on reliability, latency reduction, cost control, and clean code over theoretical perfection.
• Strong Collaborator: Comfortable working alongside backend engineers, data teams, and product owners in a fast-paced environment.

Location: LATAM

USD PAY