Data Scientist – Generative AI

Argus Media; São Paulo; Hybrid; BRL 180,000 - 320,000; Full time

Job summary

Argus Media is seeking a Data Scientist to join its Generative AI team. You will craft GenAI applications, make Argus content AI-ready, and build retrieval and agentic capabilities to power AI-enabled products.

Your work will span transforming data, developing semantic/vector retrieval, and creating agentic workflows with LangChain and LangGraph, while integrating LLMs with data and external services.

Qualifications

• Hands-on experience building GenAI applications.
• Experience with LangChain and LangGraph to build LLM-powered workflows.
• Strong in retrieval, semantic search and embeddings.
• Experience integrating LLMs with data, tools or external services.
• Experience preparing structured or unstructured data for AI use.
• Ability to evaluate LLM, retrieval or agentic systems.
• Cloud experience (Azure/AWS/GCP) and Docker experience.
• Strong problem-solving, communication and collaboration skills.

Responsibilities

• Develop customer-facing GenAI capabilities across Argus products.
• Transform and structure content and data for AI readiness.
• Advance semantic, vector and hybrid retrieval across data sources.
• Build agentic apps and workflows using LangChain and LangGraph.
• Develop APIs and tooling to enable reliable AI interactions with data.
• Evaluate and optimize GenAI systems and components.
• Deliver high-quality Python services and data pipelines.
• Collaborate with cross-functional teams to scale GenAI solutions.

Skills

Python

LangChain

LangGraph

RAG

Semantic search

Embeddings

Vector databases

Retrieval techniques

LLM integration

Data preparation

LLM evaluation

Azure

AWS

Google Cloud

Docker

Problem-solving

Communication

Collaboration

Tools

Azure

AWS

Google Cloud

Docker

Job description

Argus is where smart people belong and where they can grow. We answer the challenge of illuminating markets and shaping new futures.

What We're Looking For

Join our Generative AI team to help shape how AI is used across Argus.

We are looking for a Data Scientist with hands‑on experience developing Generative AI applications. Working alongside other data scientists and engineers, you will help develop customer‑facing GenAI solutions and the data capabilities that underpin them.

A key initial focus will be helping make Argus content and data AI‑ready, and developing the retrieval, evaluation and agentic capabilities needed to use that data effectively in AI‑powered products.

You will also contribute to the wider development of GenAI capabilities as our technology and use cases evolve.

What will you be doing?

As Part Of a Multidisciplinary GenAI Team, You Will

• Customer‑facing GenAI: Help develop and improve GenAI capabilities used in Argus products, combining LLMs, retrieval, data and agentic approaches to solve real customer problems.
• AI‑Ready Data: Transform, enrich and structure Argus content and data so that it can be effectively understood and used by AI systems.
• Search & Retrieval: Develop and improve semantic, vector and hybrid retrieval capabilities across Argus content and data.
• Agent Development: Develop agentic applications and workflows using LangChain and LangGraph, combining tools, retrieval and LLM reasoning.
• Agent Tooling: Help develop APIs, MCP tools and other reusable capabilities that allow AI agents to interact reliably with data and services.
• Evaluation & Optimisation: Evaluate and improve the quality and reliability of GenAI systems, including retrieval, tools and agent behaviour.
• Technical Delivery: Develop high-quality Python components, services and data pipelines, applying good software engineering practices.
• Collaboration: Work closely with data scientists, engineers, editorial, product and business teams to develop scalable GenAI solutions.

Skills And Experience

• Strong Python skills, with hands‑on experience developing Generative AI applications.
• Practical experience using LangChain and/or LangGraph to build LLM‑powered applications or agentic workflows.
• Practical experience with RAG, semantic search, embeddings, vector databases and modern retrieval techniques.
• Experience integrating LLMs with data, tools or external services.
• Experience working with and preparing structured or unstructured data for effective use by AI systems.
• An understanding of approaches for evaluating and improving LLM, retrieval or agentic systems.
• Practical experience with cloud platforms such as Azure, AWS or Google Cloud and technologies such as Docker.
• Strong problem‑solving, communication and collaboration skills.

Experience with Model Context Protocol (MCP), GraphRAG or knowledge graph‑based retrieval, entity resolution or LLM evaluation would be an advantage.

Argus is the leading independent provider of market intelligence to the global energy and commodity markets. We offer essential price assessments, news, analytics, consulting services, data science tools and industry conferences to illuminate complex and opaque commodity markets.

Headquartered in London with 1,500 staff, Argus is an independent media organisation with 32 offices in the world’s principal commodity trading hubs.

Companies, trading firms and governments in 160 countries around the world trust Argus data to make decisions, analyse situations, manage risk, facilitate trading and for long‑term planning. Argus prices are used as trusted benchmarks around the world for pricing transportation, commodities and energy.

Founded in 1970, Argus remains a privately held UK‑registered company owned by employee shareholders and global growth equity firm General Atlantic.