AI Agent Developer

MedVirtual; Philippines; On-site; PHP 1,200,000 - 2,400,000; Full time

Job summary

MedVirtual is seeking an experienced AI Agent Developer to research, design, develop, and manage software applications end-to-end. The role focuses on building production-grade autonomous AI agents and deploying them at scale in real-world environments.

You will implement RAG architectures with vector databases, integrate LLMs via APIs, and ensure security with MCP, AI governance, and safety tooling. Collaboration with engineers, designers, and analysts is essential as you optimize performance

Qualifications

• 4+ years in AI/ML engineering or advanced data science.
• Experience shipping autonomous AI agents into production.
• Strong grounding in context engineering.
• Hands-on experience with agent frameworks: LangChain, LangGraph, CrewAI, AutoGPT.
• Experience building RAG systems on vector databases.
• Strong Python skills and fluency with AI/ML libraries.
• Experience wiring LLMs into applications via API.
• Knowledge of AI governance and evaluation frameworks.
• Familiarity with data privacy requirements.

Responsibilities

• Build and deploy production-grade autonomous AI agents.
• Design and implement RAG architectures backed by vector databases.
• Develop and optimize multi-agent workflows and agentic systems.
• Integrate large language models via API — OpenAI, Hugging Face, Azure AI.
• Implement Model Context Protocol (MCP) for secure data access.
• Own AI governance and model lifecycle management.
• Apply AI safety controls and risk mitigation measures.
• Collaborate with developers, UX designers, and analysts.
• Continuously optimize LLM performance and inference costs.

Skills

Python

AI/ML engineering

Autonomous AI agents

RAG architectures

Vector databases

LLM integration via API

Model Context Protocol (MCP)

AI governance

Safety tooling

Data privacy

Tools

LangChain

LangGraph

CrewAI

AutoGPT

Job description

As an AI Agent Developer, you will research, design, develop, and manage software applications end-to-end. Your primary focus will be on building and deploying production-grade autonomous AI agents, ensuring they are not just prototypes but fully functional systems.

• Build and deploy production-grade autonomous AI agents.
• Design and implement RAG architectures backed by vector databases.
• Develop and optimize multi-agent workflows and agentic systems.
• Integrate large language models via API — OpenAI, Hugging Face, Azure AI.
• Implement Model Context Protocol (MCP) for secure data access.
• Own AI governance and model lifecycle management.
• Apply AI safety controls and risk mitigation measures.
• Collaborate with developers, UX designers, and analysts.
• Continuously optimize LLM performance and inference costs.
• 4+ years in AI/ML engineering or advanced data science.
• Experience shipping autonomous AI agents into production.
• Strong grounding in context engineering.
• Hands-on experience with agent frameworks: LangChain, LangGraph, CrewAI, AutoGPT.
• Experience building RAG systems on vector databases.
• Strong Python skills and fluency with AI/ML libraries.
• Experience wiring LLMs into applications via API.
• Knowledge of AI governance and evaluation frameworks.
• Direct experience implementing Model Context Protocol (MCP).
• Familiarity with AI guardrails and safety tooling.
• Understanding of data privacy requirements.
• Preferred: Experience working in a law firm or legal tech environment.
• Preferred: Hands-on experience with AI tools such as LLMs, automation platforms, or analytics platforms.
• Preferred: Background in marketing automation or lead generation systems.
• Preferred: Experience with chatbots or client intake systems.
• Preferred: Familiarity with document automation or legal research tools.
• Preferred: Basic understanding of APIs or no-code/low-code tools.
• Preferred: Experience building dashboards like Power BI or Tableau.
• Preferred: Knowledge of data privacy and compliance considerations.
• Preferred: Prior experience in a consulting or operations role.