About the role
About the Role
We are a fast-growing SaaS company building AI-powered lending software. We are looking for a hands-on, curious Junior Agentic Developer to build and test AI agent workflows on our platform, and to help onboard new customers as they migrate onto it.
Working directly with our CTO, you will design autonomous and human-in-the-loop agents that automate core lending operations for our customers, including loan applications, KYC verification, credit checking, collections and loan disbursement. You will rapidly build pilots, measure their accuracy and safety, and help move successful workflows into production.
What You Will Do
• Build Agentic Pilots: Design, prototype and iterate on multi-step AI agent workflows for loan applications, document verification, credit risk checks, collections and disbursement triggers.
• Integrate with Core Systems: Connect AI agents securely to our Python (FastAPI) backend, PostgreSQL databases and third-party APIs (e.g., KYC / e-KYC providers, credit bureaus, payment gateways) using tool calling and the Model Context Protocol (MCP).
• Build Across the Stack: Contribute features to our backend and our React / TypeScript web app, with automated tests for every change.
• Deploy & Monitor: Package agent services with Docker and deploy them on Google Cloud Platform (GCP), setting up logging, tracing and guardrails to track latency, costs and decision accuracy.
• Implement Guardrails & Evaluations: Build deterministic checks and human-in-the-loop approval steps, ensuring agents never hallucinate financial decisions or trigger unauthorised disbursements.
Customer Onboarding & Migration
• Data Migration: Help move each new customer’s data from their legacy systems onto our platform using repeatable import scripts, and check that every record and balance is complete and accurate.
• Setup & Go-Live: Set up each customer on our platform, including their users, settings and workflows, and support them through go-live.
• Training & Support: Train customers’ staff remotely or on site, prepare short how-to guides, and be a first point of contact as they settle in.
Tech Stack & Requirements
Core Backend & Cloud (Must-Have)
• Languages & Frameworks: Solid proficiency in Python and experience building REST APIs with FastAPI (or Flask / Django).
• Databases & SQL: Strong SQL and PostgreSQL skills; experience migrating data between databases (e.g., SQL Server, MySQL) is a strong plus.
• Cloud & Containers: Hands-on familiarity with Docker, Linux and Google Cloud Platform (GCP) services such as Cloud Run, Cloud SQL, Secret Manager and IAM.
• AI & LLM APIs: Practical experience building with Anthropic Claude (via the Anthropic Python SDK, Claude Agent SDK, or Claude on Google Cloud Vertex AI), including structured outputs and tool / function use.
Agentic & Workflow Engineering (Strong Plus / Willingness to Learn Fast)
• Orchestration: Experience with agentic or workflow tools such as LangGraph, Google Agent Development Kit (ADK), or custom Python state machines and job queues.
• Protocols & Tooling: Familiarity with Model Context Protocol (MCP) for connecting agents to internal tools and databases, plus basic prompt engineering and evaluation techniques.
• Frontend: Working knowledge of React and TypeScript.
• DevOps Basics: Comfort with Git, Docker Compose and automated testing (pytest).
Mindset & Background
• 2+ years of software engineering experience (fresh graduates with strong personal / hackathon AI agent projects are strongly encouraged to apply).
• Strong sense of ownership, security and caution, understanding that in fintech and lending, data privacy and financial accuracy are non-negotiable.
• Patient and clear when explaining technical topics to non-technical users, and willing to travel occasionally for customer onboarding (company-paid).
• Based in Jakarta, Indonesia, with strong communication skills in Bahasa Indonesia and working proficiency in English (for technical documentation, prompts and customer communication).
Why Join Us?
• Direct Executive Visibility: Work directly with the CTO on high-priority strategic projects rather than getting lost in layers of management.
• Real-World Fintech Impact: See your prototypes evolve into live systems that our customers rely on every day.
• Modern AI Stack: Dedicated budget and access to top-tier Anthropic models and GCP infrastructure to experiment and learn rapidly.