Back-end Developer, Java
Posted 9hrs ago
Employment Information
Report this job
Job expired or something wrong with this job?
Job Description
Desenvolvedor Backend Java construindo sistemas bancários e agentes de IA para fintech brasileira. Integrando LLMs, microsserviços e automações seguras em ambiente remoto.
Responsibilities:
- Design, develop and maintain high-availability banking systems in Java and Quarkus, using a microservices architecture and messaging (SQS/SNS)
- Design and implement single-agent and multi-agent AI agents to perform business tasks
- Orchestrate tool calls, define scopes of action, security policies and autonomy limits
- Integrate LLMs via API into backend systems, using RAG, function calling, context memory and MCP
- Apply prompt engineering for code generation and agent behavior design
- Review and debug code and decisions generated by agents/AI, identifying logical errors, hallucinations and edge cases
- Lead custom implementations when automation is insufficient
- Design and validate PoCs and MVPs of agentic solutions with business stakeholders
- Build observability and evaluation for agents in production, including logging, tracing and metrics
- Identify security weaknesses and implement safeguards, including output validation and human-in-the-loop controls
- Develop unit and integration tests using Maven and Git
- Work with product, technology and business teams to translate operational pain points into scalable agentic solutions
Requirements:
- Degree completed or in progress in Computer Science, Software Engineering or related fields
- Strong backend experience with Java and Quarkus, or Spring with willingness to migrate
- Practical experience building AI agents or applications with LLMs
- Knowledge of tool orchestration, RAG, memory and agent evaluation
- Knowledge of MCP and function calling/tool use with APIs such as Claude or OpenAI
- Mastery of prompt engineering and agent guardrails
- Experience with relational databases MySQL and PostgreSQL
- Experience with NoSQL databases Redis and MongoDB
- Preferably experience with vector databases such as pgvector, Pinecone or Weaviate for RAG
- Hands-on experience with event-driven architecture using AWS SQS and SNS
- Proficiency in unit and integration testing, Git, Maven, agile methodologies and CI/CD pipelines
- Ability to read technical documentation in English
- Good communication in Portuguese or English
- Autonomy and proactivity in a 100% remote environment
- Differentials: observability of LLMs/agents, open-source contributions related to LLM tooling and experience in fintech/regulated banking systems
















