AI Engineer - Risk & Collections with Claude

Konfío; Quintana Roo; On-site; MXN 900,000 - 1,300,000; Full time

Benefits offered by this job

Impact at Scale

AI Innovation

Culture of Excellence

Flexible Work

Job summary

Konfío is leveraging Claude to build risk and collections AI agents that publish evidence-backed findings. You will optimize parallel processing and reduce costs while ensuring auditable access for external agencies.

The role entails tight collaboration with the Risk Data Science team to validate AI findings and continuously improve evaluation frameworks. You will implement robust authentication controls (OIDC/Okta) and work within AWS, Git, and CI/CD to deliver scalable, compliant AI solutions

Qualifications

• Experience with Claude and advanced prompt engineering for production-grade AI agents.
• 3+ years deploying Python in production environments and strong SQL skills.
• Experience implementing authentication/authorization (OIDC/Okta) in web apps.
• Solid understanding of Git, CI/CD (GitLab CI), and AWS at a developer level.
• Nice-to-have: Credit/Collections background; familiarity with Claude Agent SDK; MySQL/Athena.

Responsibilities

• Develop and maintain AI agents for Risk and Collections with structured, evidence-backed outputs.
• Improve parallel processing and optimize LLM costs via caching, batching, and routing.
• Ensure reliability of daily processing with retries and strong evals/prompts versioning.
• Implement strict access controls for external agencies with per-user audit trails.
• Collaborate with Risk Data Science to validate AI findings and incorporate feedback.

Skills

Claude/LLM expertise

Python production

SQL

Security/OIDC/Okta

AWS / DevOps

Git / CI/CD

Tools

Claude SDK

Athena

MySQL

GitLab CI

AWS IAM / Secrets

Job description

Konfío is leveraging Claude to build risk and collections AI agents that publish evidence-backed findings. You will optimize parallel processing and reduce costs while ensuring auditable access for external agencies.

The role entails tight collaboration with the Risk Data Science team to validate AI findings and continuously improve evaluation frameworks. You will implement robust authentication controls (OIDC/Okta) and work within AWS, Git, and CI/CD to deliver scalable, compliant AI solutions