AI, with less. Mess to insight. Days to seconds. Bucks to pennies.
I earned my Ph.D. in Computer Science from
UNSW
(1995) and am now a Full Professor at
NC State University
, where I direct the
Irrational Research Lab
. My work focuses on
software engineering for AI
, specifically building
data-driven, explainable, and intelligent software systems
. I am an
ACM, IEEE, and ASE Fellow
, and co-creator of the
PROMISE repository
, helping establish modern empirical software engineering by showing that
small, interpretable AI models
can often outperform larger, more complex ones.
I have published over
300 papers
with more than
24,000 citations
(h-index=74) and have advised
24 Ph.D. students
. I serve as the
Editor-in-Chief
of the
Automated Software Engineering
journal and, from
2010 to 2026
, as an
Associate Editor for IEEE TSE
.
My research is supported by
$19M
in competitive grants from government (
NASA
) and industry (
LexisNexis, Microsoft, Meta
). Recent highlights include over
$2.5M from the NSF
and
$1M from the NSA
to advance the science of trustworthy and compact AI. For more information, visit
timm.fyi
.
to help me find and
fix problems in
real world AI/ML systems. Are you are NCState student, or an industrial
partner, interested in research? Why
not join my
research
reading group (run each semester)?
Research
AI, for Less
I strongly advocate for "AI, for Less". My research shows
you do not always need massive computational power —
simpler, optimized models can yield exceptional results while remaining
transparent and trustworthy. We emphasize reproducibility and
simplicity over "black box" deep learning, and show that
lightweight active learning, co-training, and warm-starts
can match or beat expensive alternatives.
- FSE'26: How Low Can You Go? The Data-Light SE Challenge [pdf]
- VERIFAI'26: From Verification to Herding: Exploiting Sparsity [pdf]
- TOSEM'26: Can LLMs Warm-Start SE Active Learning? [pdf]
- CACM'25: The Case for Compact AI [pdf]
- TOSEM'24: Learning from Very Little Data
- EMSE'24: When Less Is More: Co-Training for Defect Prediction
- IEEE Access'24: iSNEAK: Partial Ordering for Model-Based Reasoning
- EMSE'22: DebtFree: Minimizing Labeling Cost in Technical Debt
LLMs & Deep Learning
Critical assessments of where LLMs and deep learning help
— and where they hurt — in software engineering tasks. We study how to
prompt, warm-start, and robustify LLM-based SE pipelines,
and identify when simpler baselines still win.
- EMSE'26: From Brittle to Robust: Improving LLM Annotations for SE Optimization [pdf]
- MSR'26: Beyond the Prompt: Domain Knowledge for LLM Optimization [pdf]
- IEEE Internet Comput.'23: A Tale of Two Cities: Variances in Robust Deep Learning
- TSE'22: Oversampling for Deep Learning Defect Prediction
- MSR'22: Improving Deep Learning for SE Analytics
- MSR'22: Dazzle: GANs for Security Class Imbalance
Security & Vulnerability
Practical methods for vulnerability detection and
cyberthreat intelligence, using active learning and adversarial
ML to help teams prioritize security efforts without drowning
in false positives.
- KAIS'25: Mining Attack Patterns from CTI Reports [pdf]
- ICDM'24: ChronoCTI: Temporal Cyberattack Patterns
- EMSE'22: Omni: Automated Ensemble with Unexpected Models Against Adversarial Evasion Attack
- EMSE'21: Better Security Bug Report Classification
Defect Prediction & Software Analytics
Building better predictors for software quality, project
health, and technical debt — using smarter sampling,
transfer learning, and hyperparameter optimization rather
than brute-force compute. Includes benchmark repositories
for reproducible research across hundreds of SE tasks.
- MSR'26: MOOT: A Repository of Many Multi-Objective Optimization Tasks [pdf]
- TSE'25: Is HPO Different for SE Analytics? [pdf]
- TSE'25: Retrospective: Static Code Mining for Defect Prediction [pdf]
- SIGSOFT SEN'25: ASE'24 Workshop: Replications and Negative Results
- ESA'23: An Expert System for Redesigning SE for the Cloud
- EMSE'22: Revisiting Process vs. Product Metrics
- EMSE'22: Predicting Open Source Project Health
- MSR'22: Stabilizing Models Across Many Projects
- ICSE'21: Early Life Cycle Defect Prediction. Why? How?
Accountability & Trust
AI software must meet the same correctness and auditability
standards as any safety-critical system. We develop methods to verify,
audit, and correct AI outputs — ensuring decisions can be
inspected, reproduced, and challenged — and probe the
fragility of causal reasoning in high-stakes environments.
- JSS'26: A Heuristic for Explaining Optimization [pdf]
- IEEE SW'25: AI in State Courts: Navigating Innovation and Ethics [pdf]
- EMSE'25: Shaky Structures: Causal Graphs in SE [pdf]
- EMSE'23: VEER: Interpretability of Model-Based Optimizations
- TSE'24: FairBalance: Equalized Odds via Pre-Processing
- IEEE Access'24: Trading Off Scalability, Privacy, and Performance
Other
- ICSE'26: SE Journals in 2036: Looking Back [pdf]
- TSE'24: Scoping SE for AI: The TSE Perspective
- IST'24: Note on the Contributions of Guenther Ruhe
- CACM'23: (Re)Use of Research Results
- IEEE Computer'22: AI and SE: Are We Ready?
