Research Internship, FlexOlmo at The Allen Institute for AI

Berkeley, California, United States

Apply Now

$94,320 – $140,000Compensation

InternshipExperience Level

InternshipJob Type

UnknownVisa

AI Research, Machine LearningIndustries

Requirements

Pursuing a PhD degree in Computer Science or similar field
Research experience in machine learning, natural language processing, language and vision, or related areas
Outstanding individual contributor (IC) skills, especially with deep learning frameworks (e.g., PyTorch)
Outstanding publication record at AI-related venues (e.g., NeurIPS, ICLR, ICML, COLM, ACL, EMNLP), evaluated for rigor and impact
Research experience in large language models, training dynamics, scaling laws, and data curation
Experience with mixture-of-experts, long-context language models, and retrieval (preferred but not required)
PhD candidate or master/undergraduate student with a strong research background

Responsibilities

Define and lead a high-impact research project
Train and release leading models
Collaborate with and learn from team members across Ai2
Build open-source software for the research community
Author scientific papers for publication in high-profile conferences or journals
Contribute to cutting-edge research in natural language processing and machine learning, focusing on Mixture-of-Experts (MoE), long-context language models (LCLMs), and retrieval

Skills

Key technologies and capabilities for this role

Mixture-of-ExpertsMoElong-context language modelsLCLMsretrievalnatural language processingNLPmachine learninglarge language modelsmodel trainingopen-source software

Questions & Answers

Common questions about this position

What is the compensation for this research internship?

Compensation is dependent on current degree enrollment: Research Intern, Undergrad: $94,320 annually; Research Intern, Masters: $106,110 annually; Research Intern, PhD: $140,000 annually.

Is remote work allowed for this internship?

Persons in these roles are welcome to work remotely from Berkeley, CA, though interns will be based in Berkeley with opportunities to engage with UC Berkeley and the BAIR lab.

What qualifications are required for this research internship?

Candidates must be a PhD candidate or a master/undergraduate student with a strong research background.

What is the team culture like for the FlexOlmo internship?

FlexOlmo is a small, tightly knit team that encourages open collaboration on intern projects, even with external researchers, in a fast-paced research environment focused on high-impact work.

What is the application deadline and process for this internship?

Apply by December 15, 2025 at 11:00pm Pacific Time to be considered for Spring/Summer 2026 internships with flexible start dates in 2026; email olmo-careers@allenai.org with questions.

The Allen Institute for AI

Advances AI through research and engineering

About The Allen Institute for AI

The Allen Institute for Artificial Intelligence (AI2) focuses on advancing artificial intelligence through research and engineering. It develops technologies that improve machine reasoning, common sense for AI, and language modeling. AI2 works on various projects, such as AllenNLP for natural language processing and EarthRanger for environmental monitoring. Unlike for-profit companies, AI2 operates as a non-profit, relying on grants and donations to fund its initiatives. This allows them to provide their research and tools to the public for free, promoting collaboration in the AI field. The goal of AI2 is to tackle significant scientific and engineering challenges to benefit society.

Seattle, WashingtonHeadquarters

2014Year Founded

N/ACompany Stage

AI & Machine LearningIndustries

201-500Employees

Benefits

Medical

401(k)

Visa Sponsorship

Time off

Catered meals & free snacks

Training & development

Tuition reimbursement

Remote & hybrid work options

Risks

Competition from OpenAI's QwQ-32B model may reduce AI2's market influence.

AI2's reliance on grants poses financial risks during economic downturns.

Open-source models may face intellectual property risks and misuse.

Differentiation

AI2 focuses on open-source models, promoting transparency and collaboration in AI development.

AI2's projects like AllenNLP and Semantic Scholar address diverse scientific and societal challenges.

AI2 operates as a non-profit, relying on grants and donations for funding.

Upsides

AI2's open-source models attract researchers, enhancing collaboration and innovation.

Tülu3 and OLMo-2 models position AI2 as leaders in open-source AI.

AI2's multimodal models meet the demand for integrated data processing in various fields.

Land your dream remote job 3x faster with AI

Try Jobo Free