companyVitol logo

Senior Machine Learning Engineer - Agentic Workflow

VitolHouston
On-site Full-time

Clicking Apply Now takes you to AutoApply where you can tailor your resume and apply.


Unlock Your Potential

Generate Job-Optimized Resume

One Click And Our AI Optimizes Your Resume to Match The Job Description.

Is Your Resume Optimized For This Role?

Find Out If You're Highlighting The Right Skills And Fix What's Missing

Experience Level

Mid to Senior

Qualifications

Bachelor’s or master’s degree in Computer Science, Engineering, or a related field.6+ years of experience in software engineering, machine learning engineering, or platform engineering. Proficient in writing production-grade Python code; experience with Claude Code or Cursor is a plus. Hands-on experience with LLM-based systems, including:LangChain / LangGraphMCPLangsmithClaude or comparable frontier modelsAWS AgentCore or similar agentic frameworksStrong understanding of RAG architectures, embeddings, and vector search. Experience designing and consuming APIs (REST and/or async/event-driven). Solid cloud engineering experience, particularly with AWS. Familiarity with fine-tuning frontier models for domain-specific knowledge. Experience with distillation, quantization, and small language models is an advantage. Proven experience deploying traditional machine learning models into production environments.

About the job

Join Vitol as a Senior Machine Learning Engineer / Platform Engineer, where you'll be instrumental in designing and developing a cutting-edge production-grade agentic workflow platform. This pivotal role merges LLM systems engineering, distributed platforms, and applied machine learning, with a strong focus on orchestration, reliability, and extensibility. Your expertise will help architect and implement agent-based workflows that seamlessly integrate large language models, retrieval systems, structured knowledge, and external APIs—engineered for robustness, observability, and real-world business applications.

  • Design and implement multi-agent and single-agent workflows utilizing orchestration patterns, context engineering, memory management, and guardrail strategies.
  • Create RAG pipelines that incorporate vector search, hybrid retrieval, and citation tracking.
  • Leverage knowledge graph-backed reasoning, including ontologies, entity resolution, and graph-based context construction.
  • Establish evaluation frameworks for assessing agent task completion correctness, quality, cost, and latency.
  • Develop and deploy machine learning models with a focus on production readiness, scalability, and performance.
  • Collaborate with data scientists to transition experimental models into robust, production-grade applications.
  • Integrate with collaboration platforms (e.g., Teams, alerting systems) to enable intelligent distribution of insights.
  • Implement and manage CI/CD pipelines for the automation of model deployment, testing, and monitoring.
  • Architect and deploy systems on AWS, utilizing compute, storage, and security services.

About Vitol

Vitol is a leading global player in energy and commodities, dedicated to producing, managing, and delivering these vital resources to consumers and industries worldwide. Beyond its core trading business, Vitol has invested over $10 billion in infrastructure globally, solidifying its position as a trusted partner in the energy sector. With a diverse clientele that includes national oil companies, multinationals, leading industrial firms, and utilities, Vitol operates from approximately 40 offices worldwide. In 2024, the company generated revenues of $331 billion. Our people are our greatest asset. We value talent and foster an environment that promotes individual growth and eliminates hierarchical barriers. Our team comprises over 65 nationalities, and we are committed to nurturing a diverse workforce. Discover more about us here. This role is based in Houston, TX - In-office 5 days a week.

Similar jobs

Tailoring 0 resumes

We'll move completed jobs to Ready to Apply automatically.