Research Intern, AI Systems

IBM

Confirmed live 2 days ago High trust

Quick summary

Work type: On-site
Location: —
Employment: Intern
Posted: 9 days ago
Freshness: Confirmed live 2 days ago

Market check

Salary context

How this pay compares to similar roles

Similar $203k

$139k most similar roles pay here $267k

This listing doesn't post a salary. Most similar roles pay $151,356–$254,750.

Based on 240 similar postings.

Employer

About IBM

IBM is a US-based global technology company providing hybrid cloud, AI, consulting, enterprise software, and IT infrastructure products and services.

IBM currently has 506 open roles on FindRole.

Listed pay typically runs $174,632–$197,165 across 5 roles with salary data.

Most-posted roles

View all roles at IBM

At a glance

TL;DR · Research Intern, AI Systems

Role Posting Log in to save

IBM Research Intern 2027: AI Systems is a research internship within the AI Native Systems organization focused on enhancing the efficiency, reliability, and scalability of agentic AI systems for IBM Z. The intern will design, implement, and evaluate adaptive information-control mechanisms to help agents manage complex, long-horizon system-management tasks. Key responsibilities include modeling minimal sufficient agent states, developing techniques for context compression and memory management, and evaluating performance metrics like inference efficiency and resource consumption on platforms such as IBM Spyre. The role requires proficiency in Python and experience with machine learning frameworks like PyTorch or Hugging Face. Candidates should possess expertise in large language models, retrieval-augmented generation, and distributed systems to solve technical challenges involving information acquisition and evidence-sufficiency estimation within enterprise infrastructure environments for next-generation AI capabilities.

Skills

Python PyTorch TensorFlow Hugging Face LLMs RAG AI Agents Natural Language Processing Distributed Systems Linux IBM Z z/OS Memory Management Information Retrieval Inference Optimization Cloud Infrastructure

What you'll do

Research and prototype adaptive information-control mechanisms for agentic AI systems in IBM Z environments.
Model minimal sufficient agent states including observations, memory, interaction history, and task progress.
Develop techniques for information acquisition, retrieval, summarization, and context compression in long-horizon workflows.
Implement experimental solutions in Python and integrate them into the Finn/Paver system-management framework.
Execute experiments on z/OS management tasks to measure reliability, memory consumption, and inference efficiency.
Analyze trade-offs between agent performance, computational cost, and resource requirements for IBM Z hardware.
Document research outcomes through technical reports, presentations, and potential intellectual property disclosures.

What we're looking for

Must hold a Master's degree or be currently pursuing a PhD in a related technical field like Computer Science or AI.
Demonstrated experience in machine learning, deep learning, large language models, or natural language processing.
Strong programming skills in Python and experience with frameworks such as PyTorch or TensorFlow.
Experience designing and evaluating experiments and developing research prototypes in AI or distributed systems.
Excellent written and verbal communication skills to document technical work and present findings to various audiences.
Preferred experience in agentic AI, retrieval-augmented generation (RAG), or long-context reasoning.
Familiarity with Linux-based environments, distributed systems, cloud infrastructure, or enterprise computing platforms.
Experience developing, fine-tuning, or evaluating AI/ML models using Hugging Face or similar tools.

Save