AI Agent Engineer Intern — End-to-End Runtime in 30+ Languages

TikTok; Singapore; On-site; SGD 17,000 - 30,000; Full time

Job summary

TikTok is seeking an AI Agent Algorithm Engineer Intern (Global E-Commerce, Conversational AI) for a 2027 start. You will design and ship end-to-end agent loops, memory systems, and tool integration across 30+ languages, contributing to real production systems from day one.

As a PhD candidate, you will apply ML/DL/NLP fundamentals to post-train and fine-tune models, build runtimes, and evaluate performance with a focus on CSAT and containment metrics.

Qualifications

• Minimum: Currently pursuing a PhD in Software Development, CS, CE, or related field.
• Strong Python plus one of C++ / Go / Rust / Java.
• Solid ML/DL/NLP fundamentals with hands-on experience with LLMs or agents.
• Able to read a paper or engineering blog and turn it into working code.
• Preferred: Built the runtime with an agent loop / memory / tool integration.

Responsibilities

• Build and evolve the agent runtime powering customer-service agents — orchestrate skills, tools, and context; implement loop control & guardrails.
• Engineer context and memory for long multi-turn agents — memory, context editing, retrieval.
• Post-train and fine-tune LLMs and build data flywheel from conversations into signals.
• Design and integrate tools, Skills, and MCP connectors; tool search for large inventories.
• Build evaluation harnesses and close offline-online gaps.
• Develop the self-evolving loop for continual improvement.
• Own an end-to-end surface and ship to production across 30+ languages.

Skills

Python

C++

Go

Rust

Java

ML/DL/NLP

LLMs/Agents

Education

Pursuing a PhD

Tools

Agent runtime

MCP connectors

Context memory

Job description

TikTok is seeking an AI Agent Algorithm Engineer Intern (Global E-Commerce, Conversational AI) for a 2027 start. You will design and ship end-to-end agent loops, memory systems, and tool integration across 30+ languages, contributing to real production systems from day one.

As a PhD candidate, you will apply ML/DL/NLP fundamentals to post-train and fine-tune models, build runtimes, and evaluate performance with a focus on CSAT and containment metrics.