Research Assistant - College of Engineering

About the role

This is an exciting opportunity for someone who thrives in an interesting and challenging work environment. In this role, you will develop and apply artificial intelligence and machine learning methods for scientific lead optimization and laboratory search. The position will focus on developing large language model (LLM) and agentic AI systems that can retrieve and synthesize scientific information, reason over experimental data, propose and prioritize candidate leads, and support iterative laboratory decision-making. The work will integrate LLMs with domain-specific data, retrieval-augmented generation, scientific databases, predictive ML models, and experiment-planning workflows to accelerate discovery while maintaining traceability, evaluation, and appropriate human oversight.

Responsibilities

• Develop LLM- and agent-based systems for laboratory search that retrieve and synthesize scientific literature, protocols, prior experimental results, and structured scientific data using RAG, embeddings, vector databases, and tool/API integrations.
• Develop AI/ML methods for lead optimization, including candidate generation, ranking, property prediction, multi-objective optimization, active learning, and uncertainty-aware prioritization of experiments.
• Build domain-adapted LLM workflows through prompt engineering, structured outputs, fine-tuning or parameter-efficient adaptation when justified, and integration with specialized predictive models and scientific tools.
• Create evaluation frameworks for scientific accuracy, retrieval quality, grounding, hallucination/error rates, robustness, reproducibility, and usefulness of AI recommendations; implement guardrails and human-in-the-loop review for consequential laboratory decisions.
• Integrate AI recommendations with iterative design-build-test-learn or related laboratory cycles, using experimental feedback to update models, refine search strategies, and improve candidate selection.
• Develop interfaces and automation that allow researchers to query laboratory knowledge, compare candidate leads, inspect evidence and provenance, and understand the rationale and uncertainty behind model recommendations.
• Work with multidisciplinary teams to curate datasets, define scientific objectives and constraints, establish benchmarks, and translate laboratory requirements into deployable AI/ML capabilities.
• Track advances in LLMs, scientific foundation models, AI agents, retrieval/search, and AI for science; prototype promising methods and recommend approaches based on measured performance rather than model novelty alone.
• Design, implement, test, and document AI/ML software and research workflows for scientific search, data analysis, prediction, and decision support.
• Analyze experimental, computational, and literature-derived datasets; develop reproducible pipelines; communicate findings to research collaborators and stakeholders.
• Evaluate models using appropriate quantitative benchmarks, error analysis, ablation studies, and validation procedures; identify limitations and opportunities for improvement.
• Collaborate with scientists, engineers, and software developers to translate research needs into usable AI-enabled tools and workflows.
• Maintain technical documentation, code quality, version control, data provenance, and reproducibility standards.
• Contribute to reports, presentations, publications, proposals, and demonstrations as appropriate.
• Other duties as assigned.

Qualifications

• Bachelor’s degree in computer science, machine learning, artificial intelligence, data science, engineering, computational science, chemistry, materials science, bioengineering, or a closely related quantitative field.
• Master’s degree or Ph.D. in a relevant technical or scientific discipline preferred
• 2 years of experience:
  • Relevant research, software, or applied machine learning experience; advanced graduate research may substitute for professional experience where permitted by the applicable job profile
  • Hands-on experience developing and evaluating ML/AI systems using Python and modern deep-learning frameworks.
• Experience with LLM applications, RAG/search systems, scientific or engineering datasets, model benchmarking, and reproducible research workflows is strongly preferred.
• Experience connecting AI systems to laboratories, simulation, or experimental workflows is desirable.
• A combination of education and relevant experience from which comparable knowledge is demonstrated may be considered.

Requirements

• Successful background check
• Relevant cloud, machine learning, data science, or laboratory/data-compliance training where applicable preferred
• Any laboratory safety or institution-specific training required for access to experimental facilities will be completed as appropriate.

Benefits

• Comprehensive medical, prescription, dental, and vision insurance
• Generous retirement savings program with employer contributions
• Tuition benefits
• Ample paid time off and observed holidays
• Life and accidental death and disability insurance
• Free Pittsburgh Regional Transit bus pass
• Access to Family Concierge Team for childcare navigation
• Fitness center access