Agent Evaluation & Evolution Machine Learning Engineer Intern

ByteDance logo

ByteDance

10,001+ employees · China, China · Software Development

Actively hiringPaidSummer 2027Software Development
Apply Now

Best applied by Nov 22


About the role#

As an intern on the Applied Machine Learning Ark team, you will contribute to the development and operation of Large Language Model service platforms. You will work on full-stack solutions that include model training, inference frameworks, and intelligent agent systems. This role focuses on the data-driven feedback loops that inform model improvements and system optimizations.

What you'll need#

  • Current enrollment in a Bachelor's degree program in Computer Science, Artificial Intelligence, Machine Learning, Data Science, or a related field.
  • A solid foundation in machine learning and deep learning.
  • Practical experience with LLM-based systems such as agents, tool calling, retrieval, or multi-agent systems.
  • Strong Python skills and experience with a mainstream ML or agent evaluation framework.
  • Demonstrated ability in research or engineering through internships, open-source work, publications, or substantial projects.
  • Preferred: Publications at top-tier ML/NLP venues, particularly in agent learning or evaluation.
  • Preferred: Experience with metric design, model-based judging, annotation, statistical analysis, or feedback-driven optimization loops.
  • Preferred: Familiarity with LLM post-training, reasoning and planning methods, continual learning, or large-scale log and trace analysis.

What you'll do#

  • Design evaluation systems for LLM-based agents, focusing on task success, tool use, reasoning quality, and reliability.
  • Create benchmarks and automated judging pipelines that integrate rule-based checks, model-based judging, and human review.
  • Examine agent execution traces and user feedback to find failure patterns and translate them into system improvements.
  • Collaborate with research, platform, and product teams to move methods into production and support the closed loop from experience to capability.

Location & details#

  • Location: Seattle, Washington, United States
  • Term: Summer 2027
  • Work Modality: On-site
  • Position Type: Full-time intern
  • Compensation: $42.75 per hour

About ByteDance

ByteDance develops platforms for content, entertainment, and commerce. The company operates in over 30 countries and maintains a user base exceeding 2.5 billion people. Its primary focus is the creation of tools for digital interaction and media discovery. The organization employs more than 10,000 people.

LinkedInWebsiteCareers emailByteDance contact channels unlock with a free account
All ByteDance internships

How to get in at ByteDance

Applying early to ByteDance gives you a distinct advantage before the applicant pool becomes unmanageable. Intern Insider sends an instant alert the moment a role matching your target is published anywhere, so you apply among the first, and early applicants get seen before the pile grows. You can avoid the black hole of the standard application queue by connecting with the people actually managing the hiring process. Intern Insider surfaces the recruiters behind the company's roles so you can reach out directly to ask about the role or a referral, which materially improves response rates.

Get alerts for new ByteDance internships

Interns who landed roles with Intern Insider

“I went from juggling multiple tabs to having everything I needed on one platform. The convenience is unmatched and worth every penny.”
Setare M., System Integration Intern

Setare M.

System Integration Intern

“Getting my first internship was the hardest. Intern Insider's exclusive job board and AI resume builder really helped me.”
Richard Y., Finance Student

Richard Y.

Finance Student

Apply to more than this one.

10,000+ open internships in one place, with the deadlines and the people to email.

Create free account
Apply by Nov 22Apply Now
Privacy

Cookies

Manage optional cookies and similar technologies. You can update these choices anytime. Read our privacy policy

EssentialKeep you signed in and support security and basic site features.Always on
AnalyticsHelp us improve the site by measuring how people use it.
AdvertisingHelp us show relevant ads and measure how well they work.