Data Engineer II

Data Engineer II

Data Engineer II with a strong focus on innovation and Artificial Intelligence. Your challenge will be to build AI-driven products that enhance our platform, developing everything from agents and chatbots to data automation systems. Here, you will combine traditional data engineering with the latest in LLMs to generate real impact on the efficiency of our teams, actively collaborating with Data Platform and Analytics Engineering.

Your challenges with us will include:

• Building AI-driven products for the data ecosystem (agents, chatbots, automatic query rewriting systems, automated model and documentation generation);
• Helping to maintain the reliability of agent systems (guardrails, validations, and observability, such as LangFuse) to handle non-determinism in production;
• Working with MCPs and tool integrations, building and evolving internal MCPs;
• Participating in technical decisions evaluating trade-offs (RAG vs LLM Wiki, heuristic vs LLM, cost vs latency vs quality), conducting comparative experiments;
• Participating in code reviews, focusing on performance, token cost, and maintainability;
• Collaborating with Data Platform and Analytics Engineering to connect platform products to the real consumption of teams.

We expect you to have:

• Python and SQL;
• Practical experience building and deploying AI products (heavy use of LLMs in daily operations: agents, model APIs, skills/plugins like those of Claude Code), not just casual use;
• Programming logic, algorithms, and data structures;
• Orchestration of pipelines (Airflow, Dagster) and data transformation with dbt (or Dataform);
• Good familiarity with cloud systems and architectures (preferably GCP/BigQuery);
• Experience with deployment (Kubernetes, Docker, Cloud Run);
• Notion of how to integrate systems (APIs, webhooks).

It will be a plus if you have:

• Agent systems in production (prompt engineering, agent orchestration, tools, sub-agents, parallelism, retry, and fallback);
• MCP, Model Context Protocol (construction or consumption of MCPs to standardize agent access to data sources);
• RAG and alternative architectures (e.g., prior synthesis style LLM Wiki) and their trade-offs;
• Evaluation of AI systems (LangFuse, quality/cost/latency metrics, design of evals for non-deterministic systems);
• Spec-driven development (Spec Kit, SDD): constitutions, quality gates, and versioned specs guiding implementation;
• Management and integration with visualization platforms (Metabase, Looker, Tableau) via API;
• Claude Code skills to automate team workflows;
• Infrastructure as code (Terraform);
• Construction of services (FastAPI) and Slack bots/integrations.

One of the pillars of Arcotech is diversity. Our motivation is to keep moving forward, innovating, and taking educational technologies to new heights. To do this, we believe in the power of equity and respect for equal rights. Here, we recognize the singularities, the differences, and we embrace everyone.

What you will find at Arco, in addition to amazing people: + Food allowance and/or meal Flash + Health and dental insurance SulAmerica + Transportation allowance for in-person and hybrid positions + Extended maternity and paternity leave + Childcare support + Health care: partnership with Wellhub and Zenklub + Incentive for education + Discount on air tickets + Partnership with pet health plan + Access to Arco teaching materials for employees' children + Partnerships for MBA and Postgraduate #MoreDiversity We believe that in a diverse and inclusive environment, our

people will be happier, more engaged

and with

more chances to reach their maximum potential

. That's why, at Arco, we respect and value equality of right, we recognize the singularities and we believe that

everyone

can contribute to our purpose.

#ComeToArco