AI Platform Engineer - 63449289357

AI Platform Engineer

Location: South Africa (expand to LATAM first, then PH, if necessary)

Employment Type: Full-Time (Monday–Friday, 9:00 AM – 5:00 PM EST)

Experience Level: 4+ Years

Salary: Maximum $4,000

Our client builds and deploys autonomous AI agents that handle core operational tasks for small and mid-sized businesses. Operating directly within customer environments, these agents answer messages, run reports, integrate with key business systems, and automate end-to-end workflows.

Each customer receives a dedicated, secure AI agent running on dedicated AWS infrastructure. As the company scales, they are looking for an experienced AI Platform Engineer to lead the deployment, orchestration, and continuous optimization of these production AI agents and the underlying core platform.

As an AI Platform Engineer, you will own the full lifecycle of customer AI agents—from initial deployment and API integrations to ongoing performance optimization, security, and observability.

This is a high-ownership, highly technical builder/operator role. You will bridge software engineering, cloud infrastructure, and production AI to turn ambiguous customer requirements into reliable, production-grade solutions. You will be responsible for scaling agent orchestration, maintaining infrastructure security, minimizing LLM compute/token costs, and ensuring peak platform uptime.

• Deploy, operate, and maintain customer-specific AI environments across AWS infrastructure.
• Safe rollout of system updates and feature releases across live production deployments.
• Enforce Infrastructure as Code (IaC) practices to guarantee secure, repeatable, and scalable environments.
• Monitor overall platform health, ensuring high availability, system resilience, and reliability.

• Architect and enhance the core AI platform that powers multi-agent workflows and orchestration.
• Improve platform observability, tracing, metrics collection, and diagnostic tooling.
• Drive long-term platform architecture, system scalability, and technical standards.

• Design and build secure API integrations with third-party customer platforms (CRMs, email providers, scheduling platforms, project management systems, and custom line-of-business apps).
• Develop maintainable API wrappers, service connectors, and MCP servers for AI agent access.
• Handle complex integration challenges, including authentication flows, webhooks, rate limiting, pagination, and data synchronization.

• Build serverless ETL pipelines and reporting infrastructure using AWS serverless tools.
• Design robust data warehousing patterns to supply AI agents with accurate operational data.
• Deliver business intelligence tools and telemetry that surface actionable operational insights.

• Implement strict least-privilege access controls and secure credential/secret management for customer agents.
• Establish robust authentication/authorization workflows and safeguards against security risks (e.g., prompt injection).
• Ensure complete data isolation and protection across customer environments.

• Optimize AI agent speed, output accuracy, and compute/token efficiency.
• Fine-tune model selection, context management, and inference routines to reduce operational costs.
• Establish cost-attribution tracking across individual customer deployments.

• Diagnose and resolve production incidents, agent failures, and integration bottlenecks.
• Conduct root-cause analyses and implement preventative safeguards to eliminate recurring issues.

• Experience: 4+ years of professional software engineering experience with demonstrated production ownership.
• Core Languages: Strong proficiency in TypeScript, Node.js, and Python.
• Data: Advanced SQL skills with experience building production-grade data pipelines and storage models.
• Cloud & IaC: Hands-on experience with AWS core services (Lambda, S3, IAM, EC2) and Terraform (or equivalent IaC tools).
• Integrations: Deep understanding of REST APIs, webhooks, OAuth/authentication flows, rate limiting, pagination, and data validation.
• AI/LLMs: Proven experience building and supporting production-level AI applications beyond basic wrapper or chatbot implementations.
• Communication: Exceptional written and verbal English communication skills, with the ability to interface with customer stakeholders and translate ambiguous requirements into technical specifications.

• Experience operating LLM-powered autonomous agents in production environments.
• Hands-on experience with agent orchestration frameworks and tools (e.g., Claude Code, Agent SDKs, LangGraph, Model Context Protocol / MCP).
• Experience building or maintaining custom MCP servers.
• Familiarity with specialized AI observability, tracing, and evaluation tools (e.g., LangSmith, Phoenix, Helicone).
• Solid understanding of AI security patterns, including prompt injection mitigation and safe agent autonomy.
• Background working within fast-paced startup, technical consulting, or customer-facing engineering roles.

• Languages: TypeScript, Node.js, Python, SQL
• Cloud & Serverless: AWS (Lambda, S3, IAM, EC2), Serverless Framework
• Infrastructure as Code: Terraform
• AI & Orchestration: LLM APIs, Agent SDKs, LangGraph, MCP
• Integrations & DevOps: REST APIs, Webhooks, Git, CI/CD Pipelines

• Direct Impact: Direct ownership of production AI systems that perform real work for real businesses every day.
• Cutting-Edge Tech: Exposure to production-grade AI agent infrastructure, serverless cloud architecture, and modern orchestration tools.
• High Autonomy: A flat, engineering-driven culture that values ownership, pragmatic problem-solving, and rapid execution.
• Growth: Opportunity to help shape the core architectural foundation of a rapidly scaling AI platform.

• Reliability: High platform availability, minimal uptime disruptions, and fast resolution of production incidents.
• Integration Delivery: Timely rollout of secure, scalable customer system connectors and API integrations.
• Cost Efficiency: Measurable reduction in per-agent infrastructure and token consumption costs.
• Performance: Reduced latent response times and improved operational accuracy of agent workflows.
• Customer Satisfaction: Successful deployment of requested agent capabilities and high client satisfaction.