Senior AI Engineer - AI Agents

About the role

3Pillar is looking for a Senior AI Engineer to design, build, deploy and run AI agents that automate real business processes across Finance, Partners and Services, Support and beyond. The engineer takes an agent from a messy process description to a governed, monitored, production system used by colleagues every day, working closely with the AI Architect and Build Lead. The primary AI interface is Claude, so the focus is on data, reliable agent workflows, tools and safe cloud execution.

Key responsibilities

• Build production AI agents end to end: discovery with the process owner, design, prompt and tool engineering, testing, deployment and ongoing operation.
• Work with business owners, product managers, architects and engineers to find problems solvable with Gen AI and Agentic AI.
• Connect agents to company data via a Snowflake data foundation and to source systems such as NetSuite, Salesforce, Zendesk, Asana and PlanHat.
• Build and maintain MCP servers, tools and Claude skills so agents are available inside Claude.
• Design data and agent pipelines in Python: retrieval, structured extraction, multi-step orchestration, human-in-the-loop approvals and scheduled runs.
• Implement agent chaining so multiple agents collaborate on complex tasks, and develop and refine prompts.
• Deploy and operate agents on AWS (Bedrock, Lambda, Step Functions or equivalent) with logging, monitoring, cost tracking and alerting.
• Write evaluations that prove an agent is accurate before go-live and keep it accurate afterwards.
• Apply security and data governance by design: least-privilege access, secrets management, PII handling and audit trails, and track token and infrastructure spend per agent.
• Maintain technical documentation, share reusable patterns, mentor junior Python engineers and advocate for AI adoption.

Required profile

• Bachelor's or Master's degree in Computer Science, Engineering, Mathematics or a related field.
• 4-7 years in software or data engineering, with at least 2 years focused on LLM and agent systems in production.
• Track record of building AI agents that run in production, able to explain their architecture, failures and fixes.
• Deep knowledge of agent design: tool use and function calling, planning and orchestration, RAG, memory and context management, guardrails and human-in-the-loop design.
• Experience evaluating LLM systems with test sets, automated evals and regression testing of prompts and models.
• Understanding of data privacy, AI ethics, Responsible AI and model governance.

Required skills

• Hands-on experience with LLM APIs (Claude preferred; OpenAI or others welcome).
• At least one agent framework or SDK such as Claude Agent SDK, LangGraph, LangChain or LlamaIndex.
• Strong Python, including FastAPI, integrations and data pipelines.
• Strong SQL and experience with a cloud data warehouse, ideally Snowflake.
• Experience deploying services on AWS (or Azure), including serverless, containers, IAM, observability and CI/CD.

What we offer

• Medical insurance and life insurance as per company policy.
• 12 days of paid vacation, sick leave and paid holidays as per local law, plus paternity, maternity, marriage, bereavement and graduation leaves.
• Christmas bonus as per local law and a saving fund.
• Flexible work environment, remote-first approach and a global team.
• Wellbeing focus, fitness offerings and mental health plans (country-dependent).
• Udemy e-learning and direct deposit of payroll.