Full-Stack Engineer (Malaysian Only)

Company Description

Hatchlabs is Johor Capital Group's (JCG) wholly owned, AI-native digital-solutions company. We build production software with small, senior teams using AI as a force multiplier — problems to products in weeks, not months. Our first flagship build is the IDEAS Platform — the operating system for JCG's investment mandate in Johor and the Johor–Singapore Special Economic Zone. It runs continuous market discovery, anchor and partner engagement, an eight-stage investment workflow, and portfolio stewardship on one shared data spine, with an agentic AI layer assisting human judgement at every step. The platform is built cloud-native on Microsoft Azure, in the Malaysia West region, for in-country data residency.

Role Description

• Data pipeline. Build the Discover module's data backbone: ingestion from open data, subscriptions and listeners into a managed lakehouse, using Azure data services (Microsoft Fabric and/or Data Factory, ADLS).
• Lakehouse. Model and transform data through bronze/silver/gold layers; own data quality, freshness and incremental loads.
• Analytics. Implement signal scoring and analytical queries that turn raw market data into ranked opportunities.
• Front end. Build the analytics-facing surfaces — the Command Centre views, dashboards and persona-specific reporting — in React / Next.js over the same spine.
• Search & retrieval. Wire the vector and search layer (Azure AI Search) that powers retrieval for the platform's AI agents, in partnership with the AI engineer.
• Discipline. Keep the analytical and operational data paths cleanly separated, and the data models portable and well-documented.

Qualifications

• Full-stack capability (React / Next.js, TypeScript, Node.js) with genuine strength on the data side.
• Strong SQL and data-modelling skills; experience building ETL/ELT pipelines and working with a lakehouse or warehouse.
• Python for data work.
• Able to reason about data quality, incremental loading, schema evolution and pipeline reliability — not just write queries.
• Fluent with AI-assisted development.

Nice-to-have

• Microsoft Fabric, Azure Data Factory, Synapse, or Databricks experience.
• Dbt or similar transformation tooling; Parquet / DuckDB familiarity.
• Azure AI Search or other vector-search experience; exposure to RAG pipelines.
• Data governance/cataloguing (Microsoft Purview).

First 90 Days, Success Looks Like

• A working ingestion-to-lakehouse pipeline feeding a live opportunity log in the Discover module.
• First signal-scoring logic running and surfaced in the Command Centre.
• Clean, documented, portable data models the rest of the team can build on.

How We Work

• Small, senior, AI-native team — you will use AI coding and agent tooling daily, and we expect you to be opinionated about where it helps and where it doesn't.
• Iterative delivery in short cycles; we ship working software each cycle and build each module behind an API before its interface.
• Azure-first and Microsoft-aligned (Entra ID, Azure DevOps/GitHub, Azure PaaS), with data resident in Malaysia and a strong audit and governance posture appropriate to a regulated capital-markets context.