How your CV stacks up
Upload your CV to see how well it fits this job role
?%
Applied AI Engineer (Document AI Team) – Trafigura
Overview
Trafigura’s Digital Transformation is revolutionising the Commodities Trading industry through the development of cutting-edge AI technologies—with the Document AI team as a critical enabler. Our mission? Unlocking data-rich proprietary documents to empower high-value process optimisation and data science applications.
We are looking for an Applied AI Engineer to grow our Document AI platform, building scalable, production-ready AI solutions to transform industrial-scale workflows. In this hands-on Individual Contributor role, you will develop Agentic AI/LLM systems tailored to Trafigura’s document-intensive operations, spanning trading, asset management, and front-office/enablingExcel analysis.
Key Responsibilities
Architecture & Development
- Develop and maintain Python-based AI applications, leveraging modern frameworks like Pydantic AI, FastAPI, and asyncio for performance and scalability.
- Build and refine document workflows using LLMs, classical NLP, and agentic AI architectures (e.g., memory systems, vector-based knowledge stores, tool-calling, and guardrails).
- Prototype, evaluate, and deploy production-grade solutions within 6 months—potential projects include:
- Agentic workflow automation for complex trading tasks, including long-running executions and seamless UI integrations.
- Development of a proprietary "market intelligence email parser" to extract actionable insights for front- and middle-office teams.
- Creation of domain-specific AI annotators for cloud-native document processing.
Engineering & Systems
- Debug model performance issues, handle edge cases, and optimise systems for reliability and business impact.
- Implement rigorous AI monitoring/observability, setting up human-in-the-loop workflows and tracking evaluation benchmarks.
- Design test-driven pipelines with emphases on CI/CD, infrastructure as code (IaC), and containerisation (AWS-centric stack preferred).
Collaboration & Stakeholder Management
- Partner with Digital Transformation to translate business requirements into production-grade AI features.
- Serve as a Centre of Excellence (CoE) for Agentic AI, supporting data science and engineering teams through mentorship and tooling.
- Engage with non-technical stakeholders—clarifying complexities and embedding solutions into Trafigura’s workflows.
Reasons to use Rodeo
Start with a chat, not a search bar
Grad scheme, placement, apprenticeship? Not sure what you want yet — that's fine. Your agent talks it through with you and turns "I have no idea" into a shortlist.
It searches the market for you
Every day your agent scans the market matching roles against what actually matters to you, not just keywords on a CV.
Only hits
No noise. No "maybe this fits." Just roles with a clear explanation of why they're right — and where to focus when applying.
Requirements
Critical Qualifications
Strong expertise in at least five of the following areas:
- 5–8+ years developing production-grade AI/ML systems (cloud-native, scalable) for commercial applications.
- Python mastery (Pydantic, FastAPI, type safety, multithreading/async). Fluent in MLOps frameworks (e.g., MLflow, Phoenix).
- Hands-on experience tuning and maintaining LLM agents, including:
- Memory systems (persistent context handling),
- Tool/knowledge integration («tool calling», vector-based lookup),
- Guardrails (context filtering, response calibration).
- Proven experience with observability tools like Prometheus/Grafana, model drift detection, and A/B testing mirrors.
- Deep familiarity with AI/ML evaluation—from baselines to fairness debugging (e.g., mitigation of bias/sparsity).
Tech Stack & Methodologies
- Microservices architectures, event-driven design, domain-driven design (DDD).
- Cloud engineering (AWS preferences: recipies, SageMaker, Athena).
- DevSecOps principles (image pinning, artifact governance).
Preferred Bonus
- Background in Financial Services (e.g., fixed income, equities, asset management). Familiarity with commodity risk data formats (json, xml) is** rewarding.
Success Mindset
Melds engineering pragmatism with business empathy—critical for a role where: ✔️ Your solutions solve real problems, not just "link learning papers" onto unrelated targets. ✔️ Reliability > Advanced novelty; your well-personalized LLM has 3x fewer edge case failures than your peers’ generic wrappers. ✔️ Your approach is reproducible, measurable, and key to a 5x growth impact.
Get help with your application
Your very own career expert that helps elevate your application to the next level.
Good indicators include:
- Comfort diagnosing the "why" of unexpected fractional drops in predicate accuracy (e.g., hallucinations over missing vector keys).
- Ability to refactor legacy ML artifacts (TFX served models) into Pydantic-friendly pipelines.
- Tangible experience with deploying Llama2/Flashflow pipelines in Habana or AIRO environments.
Work Environment
The Document AI team operates at the nexus of 5,000+ end users across Trafigura’s:
- Trading floor (real-time diagnostics, "what-if" trading tools).
- Risk/Compliance (automated LOB and AGR flagging).
- Data & Analytics (uncouched dashboards from unstructured data).
A Centre of Excellence with lean-budget cross-team mentorship programs supporting engineers, data scientists, and analysts.
Feedback Culture
Absolute commitment to open convos, whether foreseeing challenges in vendor vendor-selected NLP cloud services (“those rate limits are insane!”) or guiding onboarding teams through KLASA.
Reporting to the Document AI Lead—opensaleb your direct venue into the company’s possibly most under-scoped comms TRACEbility robots.
From Trafigura
We offer a dynamic global team (Luxembourg-headquartered) with Kayak-inspired co-working philosophies. Embracing： 🔹 TJA (A Year of the Same Job? Never): Leverage cross-team shadowing rota with Quant Dev or Legal AI. 🔹 Bank-Level Research Needs Covered: Onsite/remote Split! Especially relevant to those handling terminal reads. 🔹 AI Diversity Growth Pledge: Mentorship.Write a review of RAGSLM for the engineering blog.
Equal Opportunity Employer & Equal Pay Pioneer
Trafigura is an Inclusive-Pr Donovan organisation—equality at birth is guaranteed regardless of legacy, color, cognition(s), or favorus. We don’t practice greenwashing without corresponding diversity equity, inclusion & belonging progress. Active applicants transcending our threshold will be advised of our direct London offices, 34+ subsidiaries, and reference integrity protocols.
“It took my CV and asked me questions relevant to understanding what kind of jobs to suggest for me. Suggestions were almost perfect. Jobs were exactly what I’ve been looking for.”
Jessica, London
Skills
Location