Intern, AI Engineer

Descripción del puesto:

Job Description

Get to Know the Team
The Tech Infrastructure (TI) team is the backbone of Grab's engineering excellence. We are pioneering the use of AI Agents and MCP to improve how Grabbers work with technical documentation. Our goal is to transform "passive" docs into "active" knowledge bases that empower every engineer to find answers and resolve issues in seconds.
This is a paid internship role. You will report to our Principal Data Scientist and be based onsite at our office in Petaling Jaya, Selangor.
Get to Know the Role
As an AI Engineer Intern, you will be at the forefront of Grab's LLM evolution. You will work on Project Techdoc MCP, a mission-critical initiative to unify fragmented documentation servers into a performance L0 support layer for the TI Support Bot.
This isn't just a research project; you will build production-grade infrastructure, optimizing vector databases, and benchmarking AI performance against established search engines.

The Critical Tasks You Will Perform
* System Unification: Consolidate multiple Techdoc-related MCP servers into a single, unified backend to serve as the primary retrieval source for the TI Support Bot.
* Infrastructure Management: Improve our pgvector storage layer and explore advanced chunking strategies to improve retrieval accuracy.
* Pipeline Engineering: Build and monitor automated CI/CD pipelines for incremental indexing of Techdocs, ensuring the AI's "brain" is always up to date with the latest GitLab merges.
* Evaluation & Benchmarking: Conduct rigorous A/B testing and performance analysis. You will use Evalshub and Datadog to benchmark the MCP against other search and existing TI Support Bot datasets.
* Feature Innovation: Develop "Auto-techdoc" features, such as automated newsletter pushing and Slack-based user query evaluation based on metadata mapping

Requerimientos del candidato/a:

Qualifications

What Essential Skills You Will Need
* Internship Duration: Can start from May 2026 onwards with a minimum duration of 3 months
* Technical Stacks: Proficiency in Python and experience with SQL (PostgreSQL/pgvector).
* AI Knowledge: Understanding of RAG, Vector Databases, and LLM frameworks (e.g., LangChain or similar).
* Engineering Mindset: Familiarity with GitLab CI/CD and the ability to write maintainable, production-ready code.
* Analytical: Comfortable working with data dashboards to interpret A/B test results and performance metrics.
* Communication: Ability to collaborate across teams to agree on metadata migrations and platform integration.

Good-to-have Skills:
* Experience building or using MCP servers.
* Familiarity with Cursor or other AI-assisted coding tools.
* Previous experience with Graph Databases or advanced search algorithms