About this role
CodeRoad needs an experienced engineer to solve critical scalability, cost, and security challenges within their growing AI ecosystem. The incoming specialist will take ownership of transforming legacy workflows into fully autonomous systems through the creation of a robust production-grade multi-agent platform.
Daily responsibilities center on designing hierarchical agent architectures, implementing dynamic context loading, and establishing rigorous evaluation harnesses to ensure safe and reliable operations. Someone with a deep background in Python-based LLM orchestration and a strong sense of technical ownership will thrive in this autonomous nearshore environment.
What you will do
- Design and deploy hierarchical multi-agent workflows using LangGraph to convert legacy single-prompt systems into modular architectures.
- Develop domain-specific sub-agents with dynamic context loading to improve latency and token efficiency.
- Create and integrate standardized tool-calling connectors via the Model Context Protocol.
- Manage LangFuse to track reasoning traces, costs, and overall LLM observability.
- Build automated evaluation and regression testing suites using PromptFlow or PromptFoo against standard benchmarks.
- Implement enterprise security guardrails and access controls aligned with OWASP Top 10 standards for LLM applications.
What you need
- Bring at least five years of professional software development experience with a primary focus on Python.
- Demonstrate a minimum of two years of hands-on experience constructing production AI agents or complex LLM workflows via LangGraph or LangChain.
- Apply strong practical knowledge of modern foundation models, open-weights models, and advanced prompt engineering.
- Showcase direct experience with LLM observability tools such as LangFuse and automated evaluation frameworks.
- Utilize familiarity with containerized cloud infrastructure including Docker and Azure alongside the Microsoft 365 Agents SDK.
- Communicate fluently in advanced written and spoken English while maintaining a high ownership mindset.
Nice to have
- Experience working with Retrieval-Augmented Generation pipelines and vector database integrations like Pinecone, Weaviate, or Qdrant.
- Familiarity with enterprise AI security frameworks, data privacy compliance, and prompt injection mitigation strategies.
- Exposure to serverless architectures and microservices deployments on cloud platforms.
Benefits
- 100% Remote work environment
- Holidays off matching local calendar standards
- Generous Paid Time Off (PTO)
- Health insurance assistance
- Competitive USD compensation
- Clear growth opportunities and continuous learning support
Skills and tech stack
Questions candidates ask
What is the remote work policy for this position?
The role offers a 100% remote work environment based in Latin America.
What compensation and benefits are provided?
The position offers competitive USD compensation alongside health insurance assistance, generous paid time off, and holidays matching local calendar standards.
What primary tools and frameworks will I use?
You will work primarily with Python, LangGraph, LangFuse, containerized cloud environments like Azure and Docker, and various foundation models.