This is a fully remote position, open to applicants in Argentina.

📋 Description

• Design and develop agentic systems utilizing cutting-edge LLMs, encompassing agent loops, tool utilization, agent capabilities, and MCP servers.

• Construct the document understanding pipeline, incorporating OCR, sophisticated image parsing, and table extraction from both scanned and poorly formatted PDFs.

• Break down intricate legal inquiries into actionable and verifiable system steps.

• Generate evaluation sets based on real contract questions with established answers.

• Assess accuracy, citation quality, and the system's capability to respond with “not stated” rather than making guesses.

• Collaborate directly with legal and commercial stakeholders to convert their inquiries into system specifications.

⛳️ Requirements

• Proficient in Python, with experience in deploying and managing production AI systems—not just prototype or notebook projects.

• Practical experience with frontier LLM APIs and a production agent SDK.

• Strong understanding of agent architecture: including agent loops, tool use, agent skills, and MCP.

• Document AI proficiency: encompassing OCR, advanced image parsing, and table extraction from scanned documents.

• Background in evaluating and testing LLM systems, focusing on hallucination control and abstention behavior.

• Knowledge of commercial clauses such as liability caps, carve-outs, indemnity, termination, auto-renewal and notice periods, assignment and change of control, survival, governing law, pricing, and escalation terms.

• Experience in creating RAG or document QA systems specifically for regulated or legal fields.