Agentic AI Systems Engineer — Production AI & Equity

Hadl.ai; Jakarta Pusat; On-site; IDR 133,920,000 - 446,400,000; Full time

Benefits offered by this job

Equity participation

Job summary

Hadl.ai in Jakarta (Kuningan) is building the AI layer for seafood supply-chain operations. You will develop and own agents that monitor data, emails and documents, flag issues to the right person and draft actions for approval.

This hands-on role requires writing code, shipping agents, and solving problems as part of a small team of senior builders. Each agent has evaluation sets, a known error rate, and a defined cost.

Qualifications

• 2+ years of software or ML engineering experience in production.
• Experience building multi-step agents: prompting, tool use, retrieval, structured output and human approval steps.
• Experience evaluating AI systems: test sets, error metrics, review loops and regression checks.
• Experience processing unstructured business data with LLMs: email, documents or transcripts.

Responsibilities

• Turn business problems into AI use cases with clear success measures.
• Build multi-step agents that monitor operations, flag risks and draft follow-ups for approval.
• Process email, documents and meeting records with LLMs: extraction, summaries and commitments logged for follow-up.
• Set up evaluation for every agent (test sets, error and approval rates, human review loops and regression checks) before it goes live.
• Build and run the AI infrastructure: model serving, LLM providers, vector stores and pipelines.
• Track quality, drift and cost in production, and report AI cost per use case.
• Connect agents to forecasts and data where they drive actions.
• Mentor and learn alongside passionate colleagues.

Skills

Python

English

Tools

TypeScript

Go

SQL

Job description

Hadl.ai in Jakarta (Kuningan) is building the AI layer for seafood supply-chain operations. You will develop and own agents that monitor data, emails and documents, flag issues to the right person and draft actions for approval.

This hands-on role requires writing code, shipping agents, and solving problems as part of a small team of senior builders. Each agent has evaluation sets, a known error rate, and a defined cost.