Research Scientist

Introduction

At IBM Research, we are the innovation engine of IBM. Exploring what's next in computing and shaping the technologies the world will rely on tomorrow. From advancing AI and hybrid cloud to pioneering practical quantum computing, we anticipate challenges and unlock new opportunities for clients, partners, and society. Working in Research means joining a team that accelerates discovery at the intersection of high-performance computing, AI, quantum, and cloud. You'll collaborate with leading scientists, engineers, and visionaries to push boundaries and turn ideas into reality. With a culture built on curiosity, creativity, and collaboration, IBM Research offers the opportunity to grow your career while contributing to breakthroughs that transform industries and change the world.

Your role and responsibilities

As a Research Scientist in Artificial Intelligence, you will lead and execute frontier AI research across modern machine learning paradigms. You will invent, analyze, and validate novel algorithms; connect theoretical insights with large‑scale experimental results; and translate research breakthroughs into real‑world, user‑facing systems.

You will work across foundation models, deep learning, reinforcement learning, and generative computing, contributing to both theoretical understanding and practical impact at scale.

Required education
Doctorate Degree

Preferred education
Doctorate Degree

Required technical and professional expertise
Key ResponsibilitiesLead Cutting‑Edge AI Research

• Define and lead high‑impact research agendas in areas such as:
  
  • Generative computing inspired agentic development for small models
  • Reinforcement learning and decision‑making under uncertainty
  • Formal algebra and applications to AI Software Development
• Drive projects from problem formulation through theoretical analysis, experimentation, and system integration.

Develop Novel Algorithms and Techniques

• Small‑Model‑Friendly Agent Harnesses: Build evaluation and execution harnesses tailored for small, efficient models, enabling reliable agentic reasoning, tool use, and planning under tight latency and compute constraints.
• Develop new generative computing (see mellea.ai) architectures for next wave of AI software development.
• Train customised models including data quality analysis, synthetic data generation, training and evaluation for small model agentic harnesses

Theory, Analysis, and Experimental Rigor

• Develop rigorous mathematical formulations and proofs to analyze algorithmic properties (e.g., convergence, generalization, robustness).
• Design statistically sound experiments, benchmarks, and ablations to validate research hypotheses.
• Measure impact on both offline metrics and end‑to‑end user experience.

Bridge Research and Real‑World Systems

• Embed research prototypes into production‑grade or large‑scale experimental systems.
• Collaborate with engineering teams to ensure research outcomes translate into robust, efficient, and usable AI solutions.

Publish and Communicate Research

• Publish and present original research in venues such as NeurIPS, ICML, ICLR, ACL etc
• Communicate insights clearly to technical and business stakeholders.

Preferred technical and professional experience

Expertise in AI Frameworks: Experience with latest RL frameworks and agentic harnesses.

* Publication in Top-Tier Communities: Experience with publishing research in top-tier technical communities, such as Neurips, ICML, ICLR, IJCAI, AAAI, etc demonstrating expertise in communicating complex research findings.

* Advanced AI Technique Development: Experience in creating and leveraging machine learning, NLP, and other AI techniques to solve complex analytical problems using rigorous and quantitative approaches.

Years of Experience:
2-4