AI Agent Engineer

Teoh Capital; Philippines; On-site; PHP 1,200,000 - 2,200,000; Full time

Job summary

Teoh Capital seeks an AI Agent Engineer to design, build, and ship production-grade AI agents that automate real workflows across products and operations. This is not a toy role—you'll own agents end to end: scoping the use case, architecting the agent, integrating it with systems, and iterating based on real-world performance.

You must be AI-native: someone who treats AI as central to how they build, using AI coding assistants daily and thinking in agentic workflows.

Qualifications

• Proven experience building and shipping AI agents or LLM-powered apps (personal projects with real users count).
• Strong software fundamentals: Python and/or TypeScript, APIs, databases, version control, testing.
• Hands-on experience with LLM APIs (Anthropic Claude, OpenAI, or similar), function/tool calling, and structured outputs.
• Experience with agent orchestration patterns—via frameworks (LangGraph, CrewAI, OpenAI Agents SDK) or from scratch.
• AI-native workflow: daily use of AI coding assistants and tooling, with measurable impact.
• Ability to scope ambiguous problems, communicate trade-offs clearly, and ship iteratively.
• Strong written communication and ability to write clear specs, prompts, and documentation.

Responsibilities

• Design and build AI agents that handle multi-step tasks — research, data processing, customer-facing workflows, internal operations, and integrations across business systems.
• Architect agent workflows: tool use, memory, planning, retrieval (RAG), guardrails, and human-in-the-loop checkpoints.
• Integrate agents with real systems via APIs, MCP servers, webhooks, and internal tooling.
• Build evaluation pipelines to measure agent quality, reliability, and cost — and iterate until production-trustworthy.
• Work with LLM providers and choose the right model, context strategy, and orchestration per use case.
• Collaborate with product and operations teams to identify automation opportunities and translate fuzzy requirements into scoped agents.
• Document agent behavior, failure modes, and operating procedures for non-engineers.
• Stay ahead of agent frameworks, model releases, and tooling and bring useful findings back to the team.

Skills

Python

TypeScript

APIs

LLM APIs

LangGraph

CrewAI

OpenAI Agents SDK

Claude Code

Cursor

Copilot

API design

Documentation

Tools

MCP servers

Vector DBs

RAG pipelines

Job description

We're looking for an AI Agent Engineer to design, build, and ship production-grade AI agents

that automate real workflows across our products and operations. This is not a \"prompt

tinkerer\" role — you'll own agents end to end: scoping the use case, architecting the agent,

integrating it with our systems, and iterating based on real-world performance.

You must be AI-native: someone who works with AI as a core part of how they build, not

someone who treats it as a side tool. You use AI coding assistants daily, you think in terms of

agentic workflows, and you're constantly experimenting with what the latest models can do.

What You'll Do

• Design and build AI agents that handle multi-step tasks — research, data processing,customer-facing workflows, internal operations, and integrations across business systems
• Architect agent workflows: tool use, memory, planning, retrieval (RAG), guardrails, and human-in-the-loop checkpoints
• Integrate agents with real systems via APIs, MCP (Model Context Protocol) servers, webhooks, and internal tooling
• Build evaluation pipelines to measure agent quality, reliability, and cost — and iterate until agents are production-trustworthy
• Work with LLM providers (Anthropic, OpenAI, open-source models) and choose the right model, context strategy, and orchestration approach per use case
• Collaborate with product and operations teams to identify high-leverage automation opportunities and translate fuzzy requirements into scoped, shippable agents
• Document agent behavior, failure modes, and operating procedures so non-engineers can safely run and monitor them
• Stay ahead of the curve on agent frameworks, model releases, and tooling — and bring what's useful back to the team

What We're Looking For

• Proven experience building and shipping AI agents or LLM-powered applications (personal projects with real users count — show us what you've built)
• Strong software engineering fundamentals: Python and/or TypeScript, APIs, databases, version control, testing
• Hands-on experience with LLM APIs (Anthropic Claude, OpenAI, or similar), function/tool calling, and structured outputs
• Experience with agent orchestration patterns — whether via frameworks (LangGraph, CrewAI, OpenAI Agents SDK) or built from scratch
• AI-native workflow: you use tools like Claude Code, Cursor, or Copilot as a daily driver and can demonstrate how AI multiplies your output
• Ability to scope ambiguous problems, communicate trade-offs clearly, and ship iteratively
• Strong written communication — much of agent work is writing clear specs, prompts, and documentation
• Nice-to-haves:
• Experience with MCP (Model Context Protocol) servers and connector ecosystems
• RAG pipelines, vector databases, and embedding strategies
• Evaluation frameworks (promptfoo, LangSmith, custom evals)
• Experience integrating with SaaS platforms (CRMs, helpdesks, payroll/HR systems, project management tools)
• Exposure to deploying agents in regulated or compliance-sensitive environments
• Prior startup or fast-moving team experience

How We Work

• We move fast and ship iteratively — a working prototype this week beats a perfect design next month
• We're AI-first as a company: everyone here uses AI tooling heavily, and this role sets the standard
• You'll have direct access to decision-makers and real business problems — no layers of bureaucracy between you and impact