Artificial Intelligence Engineer

Confidential Careers; México; On-site; MXN 600,000 - 1,200,000; Full time

Job summary

Confidential Careers seeks a highly skilled AI Engineer to architect, build, and deploy production-grade AI agents for Finance, Support, and Operations. You will integrate with Snowflake, Salesforce, NetSuite, Zendesk, and PlanHat, deploying securely in the cloud to power Claude-powered interfaces.

You will lead end-to-end development, design agent orchestration, and mentor junior engineers while ensuring governance, security, and cost controls.

Qualifications

• Bachelor’s or Master’s degree in Computer Science, Engineering, Mathematics, or related field.
• 4–7 years in software or data engineering, with at least 2 years building/deploying LLMs in production.
• Proven portfolio of shipped AI agents; discuss architecture/troubleshooting/scaling.
• Deep knowledge of agent design patterns: tool use, orchestration, RAG, memory, guardrails, human-in-the-loop.
• Experience with LLM APIs (Claude preferred) and AI frameworks; know when to custom-solve.
• Advanced Python programming for APIs (FastAPI) and robust data pipelines.
• Strong SQL, cloud data platforms (Snowflake preferred).
• Cloud infrastructure on AWS/Azure, serverless, containers, IAM, CI/CD, observability.
• Experience in LLM evaluation, automated evals, regression testing for prompts/models.
• Foundation in data privacy, Responsible AI, governance.
• Excellent communication to translate AI concepts to tech and non-tech stakeholders.

Responsibilities

• Lead end-to-end lifecycle of production AI agents—from discovery with business owners to design, deployment, and operations.
• Partner with product managers, engineers, and leaders to identify high-impact AI opportunities.
• Connect AI agents to Snowflake and core systems (Salesforce, NetSuite, Zendesk, Asana, PlanHat).
• Build and maintain Model Context Protocol servers and LLM skills to embed agents into existing platforms.
• Design complex data and agent pipelines in Python, including RAG/retrieval, scheduling, and multi-step orchestration.
• Implement agent-chaining architectures for multi-agent collaboration to solve complex problems.
• Refine prompts and evaluate agent accuracy; ensure reliability pre- and post-launch.
• Deploy scalable agents on AWS (Bedrock, Lambda, Step Functions) with logging and cost-tracking.
• Enforce security standards: least-privilege, secrets management, PII safeguards, audit trails.
• Track token usage and infra spend to report ROI on deployments.
• Produce technical documentation and mentor junior Python engineers.

Skills

Python

LLM/Agent design

AWS

Cloud platforms

FastAPI

SQL

Data privacy

Claude/Claude API

LangChain

Agent orchestration

Security

Education

Bachelor's or Master's in CS/Engineering

Tools

LangGraph

LangChain

LlamaIndex

Claude Agent SDK

Snowflake

Salesforce

Zendesk

dbt

Pinecone

Job description

We are a forward-thinking organization seeking a highly skilled AI Engineer to architect, build, and deploy production-grade AI agents that will automate complex business workflows across our Finance, Support, and Operations divisions.

In this role, you will transform high-level process descriptions into secure, fully governed, and continuously monitored AI systems used daily across the enterprise. Working closely with our engineering and architecture leadership, your core focus will be the backend orchestration: seamlessly connecting data sources, building reliable agentic workflows, exposing them as robust tools, and deploying them securely in the cloud to power our primary conversational AI interfaces (with a strong focus on Claude).

Key Responsibilities:

• End-to-End AI Development: Lead the complete lifecycle of production AI agents—from initial discovery with business owners to system design, prompt/tool engineering, deployment, and ongoing operations.
• Strategic Collaboration: Partner with product managers, software engineers, and business leaders to identify high-impact operational challenges ripe for Generative and Agentic AI solutions.
• Systems Integration: Connect AI agents to our foundational data warehouse (Snowflake) and core enterprise systems (e.g., Salesforce, NetSuite, Zendesk, Asana, PlanHat).
• Workflow Integration: Build and maintain Model Context Protocol (MCP) servers and LLM skills to embed agents directly into the platforms where our teams already work.
• Pipeline Architecture: Design complex data and agent pipelines in Python, encompassing RAG/retrieval, structured extraction, scheduled executions, multi-step orchestration, and human-in-the-loop approvals.
• Agent Orchestration: Implement advanced agent-chaining architectures so multiple AI agents can collaborate to solve multifaceted problems.
• Optimization & Evaluation: Refine prompts to maximize LLM accuracy and performance. Design robust evaluation frameworks to prove agent reliability before go-live and ensure continuous accuracy post-launch.
• Cloud Deployment: Deploy scalable agents on AWS (utilizing Bedrock, Lambda, Step Functions) while ensuring proper logging, observability, alerting, and cost-tracking.
• Security & Governance: Enforce strict security standards by design, including least-privilege access, secrets management, PII safeguarding, and comprehensive audit trails.
• Cost Management: Track LLM token usage and infrastructure spend to accurately report the net ROI of each deployed agent.
• Mentorship & Documentation: Produce clear technical documentation, establish reusable design patterns, and mentor junior Python engineers while championing AI adoption across the team.

Minimum Qualifications:

• Bachelor’s or Master’s degree in Computer Science, Engineering, Mathematics, or a related quantitative field.
• 4 to 7 years of software or data engineering experience, with at least 2 of those years exclusively focused on building and deploying LLMs and agent systems in production environments.
• A proven portfolio of shipped AI agents. (You should be prepared to discuss your architectural decisions, troubleshooting processes, and scaling strategies).
• Deep practical knowledge of agent design patterns, including tool use/function calling, orchestration, RAG, context/memory management, guardrails, and human-in-the-loop workflows.
• Hands-on expertise with LLM APIs (Anthropic Claude strongly preferred; OpenAI/others welcome) and major AI frameworks (e.g., LangGraph, LangChain, LlamaIndex, Claude Agent SDK), along with the engineering judgment to know when a custom solution is better than a framework.
• Advanced Python programming skills, specifically in building APIs (FastAPI) and robust data pipelines.
• Strong SQL proficiency and hands-on experience with cloud data platforms (Snowflake preferred).
• Solid cloud infrastructure experience on AWS (or Azure), covering serverless architecture, containers, IAM, CI/CD, and observability.
• Experience in LLM evaluation, including automated evals, test sets, and regression testing for prompts and models.
• A strong foundation in data privacy, Responsible AI, and model governance.
• Excellent communication skills with the ability to translate complex AI concepts to both technical peers and non-technical business stakeholders.

Preferred/Additional Experience:

• Hands-on experience with AWS Bedrock in production or AWS AIDLC methodologies.
• Prior experience developing MCP servers or native Claude skills.
• Familiarity with Snowflake Cortex, dbt, or semantic layer data modeling.
• Proficiency in TypeScript in addition to Python.
• Track record of automating processes specifically within Finance, Operations, or Customer Support.
• Direct integration experience with ERP/CRM tools like NetSuite, Salesforce, or Zendesk.
• Experience working in heavily regulated or data-sensitive sectors (Healthcare, Financial Services, Legal).
• Familiarity with vector databases (Pinecone, FAISS) and the Hugging Face ecosystem.
• Contributions to the broader tech community via open-source projects, research papers, or technical blogs.