3 days ago
Remote, Chile +4 moreSenior
Responsibilities
- Design, develop, and maintain production LLM applications for real business workflows.
- Build agentic systems that interact with tools, APIs, and external services with controls over their actions.
- Choose between agents, deterministic workflows, or hybrid approaches based on business and technical constraints.
- Design and implement RAG solutions covering chunking, embeddings, retrieval, reranking, and context management.
- Implement evaluations for response quality, information retrieval, and agent behavior.
- Identify and resolve hallucinations, execution errors, latency, token usage, inference cost, and reliability issues.
- Define validation, error handling, and monitoring mechanisms for deployed AI solutions.
- Collaborate with product and engineering teams to translate business requirements into viable technical solutions.
- Participate directly in architecture, implementation, and continuous improvement decisions.
Requirements
- Practical experience building and operating production LLM applications beyond prototypes or proof-of-concepts.
- Direct experience making architecture and implementation decisions for agent-based solutions.
- Experience with tool calling, API integrations, and external services.
- Practical knowledge of context, state, and memory management in LLM applications.
- Experience implementing production RAG pipelines, semantic search, document retrieval, and retrieval relevance improvements.
- Experience evaluating LLM application quality and agent behavior, and mitigating hallucinations, errors, and inconsistent responses.
- Experience diagnosing latency, token consumption, inference costs, and production reliability issues.
- Solid Python development experience, including backend and language-model integrations, testing, error handling, and application maintenance.
- Experience with agent orchestration frameworks such as LangGraph, Pydantic AI, or similar is desirable.
- Experience with Model Context Protocol, vector databases, semantic-search tools, observability, or LLM evaluation tools is desirable.
- Knowledge of Machine Learning and Data Science is desirable.
- Fintech, financial services, payments, fraud, regulated-process, sensitive-information protection, or AI security experience is desirable.
- Demonstrated technical ownership, architectural judgment, critical thinking, communication, autonomy, and responsibility for real user-facing LLM or agent projects.
Benefits
- Remote/home-office work arrangement.
- Monday-to-Friday schedule with hours adjusted to the assigned project cell.
- Birthday day off.
- Major medical insurance in Mexico.
- Life insurance in Mexico.
- Access to courses and certifications.
- English classes.
- Opportunities across the company’s different business lines.
- Integration with global brands and disruptive startups.
Tech Stack
Categories
About DaCodes
DaCodes is a software consultancy that designs and builds custom web and mobile applications, CRMs, and enterprise systems for businesses. The company operates on a project-based services model, providing product management, design, engineering, and ongoing support from a distributed team. Founded in 2014 and headquartered in Houston, Texas, it is privately held and serves clients across varied industries.