Introduction

IBM Research is at the forefront of utilizing AI in business settings. Our mission is to push the boundaries of generative AI technologies by harnessing the power of large language models (LLMs) and AI agents. We are developing innovative tools and technologies for GenAI application developers. As AI agents become increasingly capable and autonomous, they bring added design complexity and require careful balancing of correctness, cost, and latency. Our tools and analytics enable developers to address these challenges in the fast-evolving ecosystem of AI agent platforms, LLMs, and tools..

Our summer internship program offers an opportunity to collaborate with IBM’s leading research team for a 3-month, full-time internship (5 days a week) at our labs in Haifa or Tel Aviv.

More about our research:

Agentic systems mark a significant advancement in using AI for high-stakes business operations. However, Agentic systems also bring with them challenges that arise from their complex and undeterministic behavior. Our mission to give Agenic systems develoers the ability to visualize, understand, and optimize their systems behavior.

Focus Areas
We develop analytics to tackle multiple challenges in agentic systems, including multi-framework and multi-agent interactions, by analyzing their runtime behavior. Our key focus areas include:

• Flow Analysis: Analytics to identify issues in task execution flows and optimize task execution flows.
• Metrics and Issue Detection: Detecting instruction violations, cycles, and consistency issues to ensure system reliability.
• Optimization Recommendations: Designing strategies to adjust task flows and balance performance, latency, and cost.

Your role and responsibilities

We are seeking a motivated intern to join our research team and contribute to advancing analysis and optimization of AI agents.

As an intern, you will:

• Develop and refine analyses of data from runtime instrumentation of AI agents.
• Test and measure the accuracy of the different developed analyses.
• Research design alternatives and potential solutions for identified issues in AI agent runs.

Locations:

• IBM Research Lab, Israel (Haifa University Campus)
• IBM Site, Hashahar Tower, Givataim (near Tel Aviv Arlozorov train station)

Required education

Bachelor's Degree

Preferred education

Master's Degree

Required technical and professional expertise

• Current M.Sc. or Ph.D. student in Computer Science, Information Systems, or a related field, with a specialization in AI.
• Demonstrated expertise in key areas such as AI agents, LLMs, NLP, and classical ML.
• Experience with building a predictive ML model: from tailoring the right model to a problem, implementation and iterative improvements, to extracting insightful analytics from the results.
• Experience with classical ML approaches and graph-based algorithms
• Experience with Python.

Please add your grade sheet to your application

Preferred technical and professional experience

• Experience in building LLM-based applications using frameworks such as LangChain, LangGraph, crewAI, and Llama Stack.

• Familiarity with telemetry standards and tools like OpenTelemetry.
• Familiarity with probabilistic graphical models or probabilistic approaches
• A publication track record in peer-reviewed conferences or journals is a significant advantage.