About the role#
Join the AI Foundations Research team to contribute to the development of artificial intelligence. You will work alongside scientists and engineers on end-to-end research, focusing on multi-modal systems, data quality, novel model architectures, and training recipes. This role bridges fundamental science and engineering within a collaborative environment.
What you'll do#
- Engage in the full research lifecycle to pioneer new advancements in artificial intelligence.
- Identify core challenges, design novel prototype solutions, and validate them through rigorous experimentation.
- Conduct foundational research in areas such as novel training algorithms, efficient fine-tuning, reinforcement learning, and inference-time scaling for multi-modal models.
- Shepherd projects from ideation to completion in small, mentored teams.
- Disseminate significant results through publications in leading conferences and patent applications.
What you'll need#
- Hands-on experience with and a deep theoretical understanding of Large Language Models (LLMs), Vision-Language Models (VLMs), and Transformer-based architectures.
- Proficiency in Python for rapid prototyping, experimental setup, and implementing large-scale machine learning systems.
- Demonstrated strength in quantitative analysis, designing experiments, and deconstructing complex research problems.
- A track record of innovation, evidenced by peer-reviewed publications or strong preprints.
- Ability to clearly articulate complex technical concepts in both writing and presentations.
- A team-oriented approach with a commitment to rigorous, reproducible, and well-documented research practices.
- Strong publication record in top-tier conferences like NeurIPS, ICML, ICLR, CVPR, or AAAI.
- Advanced expertise in ML frameworks, specifically PyTorch, and full-cycle development of algorithms and systems.
- Hands-on experience with generative AI and multimodal models, from training to testing.
- Demonstrated ability to present complex research and build high-impact technical demonstrations.
Location & details#
- This is a full-time, paid internship.
- The role is on-site at IBM locations in Yorktown Heights, New York; San Jose, California; or Cambridge, Massachusetts.
- The internship term operates on a rolling basis.
About IBM
IBM provides IT services and consulting to a global client base. The company develops technology solutions across areas such as cloud computing, security, analytics, and infrastructure. It operates as a public organization with more than 10,001 employees. The corporate headquarters are located in Armonk, New York.
How to get in at IBM
Applying early at IBM gives you a measurable advantage because recruiters review applications as they arrive. Intern Insider sends an instant alert the moment a role matching your target is published, so you can submit your materials before the pile grows. Being among the first candidates seen helps you stand out in a large applicant pool. You can also improve your response rates by reaching out to the recruiters behind the roles directly. Intern Insider surfaces the specific contacts for IBM positions so you can ask about the role or a referral. This direct approach often works better than sending your resume into a queue.
