Junior Antitrust LLM Engineer

About Concurrences

Concurrences is the global platform for competition law, connecting antitrust professionals across 150+ countries for over 20 years. Lawyers, economists, judges, regulators, and academics rely on Concurrences to shape competition policy. We pride ourselves on scientific rigor, editorial quality, and continuous innovation, particularly now with our focus on generative AI. We foster a global community through high-level dialogue and the sharing of best practices.

The Hook: Junior Antitrust LLM Engineer

As a Junior Antitrust LLM Engineer, you will teach an AI system to reason about complex antitrust law, supported by an esteemed editorial board including former FTC Chair William Kovacic and former EU Commissioner Mario Monti. Your role is critical in evaluating RAG outputs, designing robust evaluation frameworks, and preventing the AI from confusing nuanced legal distinctions, like EU merger control with Brazilian CADE enforcement. Your profound legal knowledge will be the key technical leverage in this pioneering role.

What You'll Actually Do

• Evaluate RAG outputs for legal correctness: Fact-check AI answers, ensuring accurate case citations and correct jurisdictional application.
• Build evaluation frameworks: Define "right" in antitrust contexts, designing rubrics for citation accuracy and reasoning soundness.
• Design and test prompts: Craft prompts that guide the AI to reason specifically like an antitrust specialist, beyond generic chatbot responses.
• Red-team from a legal perspective: Identify scenarios that challenge or "break" the AI, exploring distinctions such as Chicago School vs. EU antitrust philosophy or cartel coordination vs. information exchange.
• Vibe code with Claude Code: Develop evaluation scripts and automate analysis, leveraging AI tools to build solutions rapidly without needing production engineering expertise.

Your core responsibility is to translate intricate antitrust law into a format the AI system can understand and reason with, then rigorously grade its performance.

A Typical Week

• Monday: Review 50 RAG outputs from the EU merger dataset. Flag systemic failures with examples.
• Tuesday: Design annotation guidelines for a new CADE dataset. Write the rules that guide the whole team.
• Wednesday: Async update to Anton with findings. Iterate on evaluation approach together.
• Thursday: The system conflates information exchange with cartel coordination. Map precedent. Test prompt variations.
• Friday: Ship two evaluation frameworks. 15 prompt variations tested. One systemic issue escalated.

Who You'll Work With

• Anton Bossenbroek (CTO): Your primary technical collaborator, with 20+ years of experience scaling AI systems, including generative AI, at top consulting firms.
• Nicolas Charbit (Founder & antitrust law expert): Guides the legal direction of your work, having built Concurrences into a definitive antitrust platform over two decades.
• Edouard (Co-founder, Operations): Manages business operations, vendor relationships, and coordinates the expert network.
• The AI Engineer (Your peer): Builds RAG pipelines; you design evaluation frameworks. You validate what they ship.

Who Will Love This Role

• Law students or recent graduates with a strong curiosity for AI and complex legal domains.
• Individuals obsessed with accuracy and adept at identifying edge cases.
• Natural structure-makers who can categorize and break down complex antitrust concepts.
• Remote-first builders thriving in an async work culture across multiple time zones.

Must-Have Qualifications

• Law degree (completed, or final year with graduation before start date).
• Solid antitrust/competition law knowledge (at least one course, thesis, or research project).
• Fluent in written and spoken English.
• Hands-on experience using LLMs (e.g., Claude or ChatGPT) for practical tasks.
• Genuine curiosity about the intersection of AI and law.

Big Pluses

• In-depth antitrust coursework, thesis, or seminar paper experience.
• "Vibe coding" experience with tools like Claude Code, Cursor, or Copilot.
• Proficiency in French (half of Concurrences' dataset is French).
• Strong prompt engineering intuition (e.g., few-shot examples, chain-of-thought).

Compensation & Logistics

This role offers a compensation of R$6,600/month (approximately €1,200/month), paid in EUR via Wise. It's an independent contractor position with monthly invoicing, starting with an initial 3-month project engagement. There is a clear path for ongoing collaboration at a higher rate.

Beyond financial compensation, you will benefit from direct collaboration with an experienced CTO in generative AI, and gain access to Concurrences' extensive 20-year institutional knowledge in antitrust law.

Interview Process

• Application: Submit your CV, optional portfolio, and three short answers.
• 15-min Intro Call with Anton (CTO): Discuss your background, interest in the role, and antitrust knowledge.
• Two 30-min Deep Dives: One focused on a real RAG failure scenario, and another on a prompt engineering walkthrough.
• 20-min Conversation with Nicolas (Founder): Explore your legal thinking and enthusiasm for the AI + law space.

Remote & Async Work Culture

Concurrences operates as a remote-first, async organization, with the CTO in Vancouver and the company based in Paris. English is required, and French is a significant advantage. Responsiveness across time zones is valued over fixed hours. The ideal candidate can start ASAP, within four weeks.

This is a unique opportunity to help define a novel role at the intersection of AI and law, shaping its future within the company and the industry.

Key skills/competency

• Antitrust Law
• Competition Law
• Generative AI
• Large Language Models (LLM)
• RAG Evaluation
• Prompt Engineering
• Legal Research
• Fact-Checking
• Evaluation Framework Design
• Legal Reasoning