巴西 - 远程：高级后端工程师（Python）（里约热内卢, RJ, BR, 29194
Brazil - Remote: Senior Backend Engineer (Python) (Rio de Janeiro, RJ, BR, 29194
我们寻找合适的人——那些希望创新、成就、成长和领导的人。我们通过投资员工并赋予他们自我发展和职业发展的能力，吸引并留住最优秀的人才。体验为全球能源行业最大的产品和服务提供商之一工作的挑战、回报和机遇。
职位机会
我们正在构建一个融合实时物理、结构化知识和自主智能的平台。我们寻找准备解决其中一个（或多个）挑战的开发者：
- 数据摄入：以最小的延迟处理大量传感器数据流。
- 智能：在稳健的多智能体系统中使大语言模型确定性和可靠。
- 上下文：建模复杂的本体论，以映射数千个物理资产。
你将加入一个高性能团队，参与核心后端开发，同时专注于以下其中一个独特方向：
核心职责
- 高性能API：使用FastAPI构建低延迟的Python服务，为前端和AI模型提供实时数据。
- 系统可靠性：调试复杂的并发问题，并确保分布式系统的生产可靠性。
- 快速交付：采用“快速交付”的理念，同时不牺牲代码质量、测试或API设计标准。
可能参与的项目路径（你可以参与一个或多个）：
流媒体与高吞吐量数据
- 构建流媒体管道：使用Kafka和Spark/Flink设计可扩展的服务，实时处理原始传感器数据。
- 时间序列优化：优化数据库模式（TimescaleDB），实现快速的历史数据检索，并实施算法检查以验证传感器读数。
GenAI与自主代理
- 构建自主代理：使用LangGraph部署有状态的代理，规划任务、查询知识图谱并执行工具，而不会产生幻觉。
- 高级RAG：构建结合语义搜索与结构化知识遍历的Graph-RAG管道，以获得有依据的答案。
图与知识工程
- 知识图谱工程：在Neo4j中设计领域本体论，定义资产、文档和时间序列数据之间的关系。
- 搜索基础设施：实现混合检索逻辑，结合向量搜索、全文搜索和图遍历。
技术栈
我们使用现代且高性能的技术栈。你应该精通核心技术，并在自己的领域有深入的知识。
查看英文原文
We are looking for the right people — people who want to innovate, achieve, grow and lead. We attract and retain the best talent by investing in our employees and empowering them to develop themselves and their careers. Experience the challenges, rewards and opportunity of working for one of the world’s largest providers of products and services to the global energy industry.
The Opportunity
We are building a platform that merges real-time physics, structured knowledge, and autonomous intelligence. We are looking for developers ready to tackle one (or more) of those challenges:
- Ingestion: Handling massive streams of sensor data with minimal latency.
- Intelligence: Making LLMs deterministic and reliable within robust multi-agent systems.
- Context: Modeling complex ontologies to map thousands of physical assets.
What You’ll Do
You will join a high-performance team, contributing to the Core Backend while focusing on one of the following distinct tracks:
Core Responsibilities
- High-Performance APIs: Build low-latency Python services (FastAPI) to serve live data to frontend and AI models.
- System Reliability: Debug complex concurrency issues and ensure production reliability in distributed systems.
- Rapid Delivery: Adopt a "deliver fast" mentality without compromising on code quality, testing, or API design standards.
Possible paths to work on the Project (you can contribute to one or more):
Streaming & High-Throughput Data
- Build Streaming Pipelines: Design scalable services using Kafka and Spark/Flink to process raw sensor data in real-time.
- Time-Series Optimization: Optimize database schemas (TimescaleDB) to enable fast historical data retrieval and implement algorithmic checks to validate sensor readings.
GenAI & Autonomous Agents
- Build Autonomous Agents: Deploy stateful agents (using LangGraph) that plan tasks, query Knowledge Graphs, and execute tools without hallucinating.
- Advanced RAG: Build Graph-RAG pipelines that combine semantic search with structured knowledge traversal for grounded answers
Graph & Knowledge Engineering
- Knowledge Graph Engineering: Design domain ontologies in Neo4j, defining relationships between assets, documents, and time-series data.
- Search Infrastructure: Implement Hybrid Retrieval logic combining Vector Search, Full-Text Search, and Graph traversal.
The Technology Stack
We use a modern, high-performance stack. You should be proficient in the Core and deeply knowledgeable in your chosen track.
- Core Backend: Python 3.12+ (FastAPI, Pydantic), Polars, Docker, Kubernetes.
- Streaming: Apache Kafka, Flink, Spark Streaming, TimescaleDB, Tiger Data.
- GenAI: LangGraph, LangChain, LiteLLM, Azure OpenAI/Anthropic/Local SLMs.
Graph/Data: Neo4j (Cypher), PostgreSQL (pg_vector, Full-Text Search).
Desired Profile
- Problem Solver: You dig into logs to find the root cause of a data spike, a silent failure, or a disconnected node in a graph.
- Modern Pythonista: You are up-to-date with modern Python async patterns and typing, ensuring code is high-performance and maintainable.
- You understand Big O notation and how to optimize code for CPU/memory efficiency.
Knowledge, Skills, and Abilities
Must Haves:
- 5+ years of experience in Python Backend development.
- Advanced English communication skills.
- Computer Science fundamentals: Data Structures and Algorithms.
- Strong experience building and documenting REST APIs (FastAPI).
Good to Have:
- Streaming: Proficiency with Streaming Technologies (Kafka, Flink, Spark) and Time-Series Databases (TimescaleDB, InfluxDB).
- GenAI: Practical experience building applications with LLMs, Agentic Frameworks (LangGraph), and Vector Databases.
- Graph: Hands-on experience with Graph Databases (Neo4j/Cypher), SQL database design, and Hybrid Search strategies
- Background in Heavy Industry, O&G, or IoT data (MQTT, OPC UA).
- Experience with Local LLMs (Ollama) for privacy-focused deployments.
- Familiarity with Data Lineage or Metadata management.
Halliburton is an Equal Opportunity Employer. Employment decisions are made without regard to race, color, religion, disability, genetic information, pregnancy, citizenship, marital status, sex/gender, sexual preference/ orientation, gender identity, age, veteran status, national origin, or any other status protected by law or regulation.
Location
Rua Paulo Emidio Barbosa 485 Q, Rio de Janeiro, Rio de Janeiro, 291941, Brazil
Job Details
Requisition Number: 209784
Experience Level: Experienced Hire
Job Family: Engineering/Science/Technology
Product Service Line: Landmark Software & Services
Full Time / Part Time: Full Time
Additional Locations for this position:
Compensation Information
Compensation is competitive and commensurate with experience.
Originally posted on Himalayas