David Nader Palacio

Seattle, WA

I am a Senior Applied Scientist @ Microsoft specializing in Artificial Intelligence and Software Research, with over 8 years of experience in machine learning and software engineering research. My expertise spans machine learning model interpretability, Large Language Model (LLM) evaluation, causal inference, and evolutionary computation, which I apply to building trustworthy, production-grade AI systems at scale.

My research explores the synergies between causal inference and deep learning to automate and improve software engineering, from interpretability methods for LLMs of code to quality assurance for agent-based systems. I have published and presented at premier venues including ICSE, TOSEM, and TSE, applying techniques such as large language models, Bayesian probabilistic methods, and information theory. My recent work includes ICSE 2026 papers on human-centered explanations of LLM-generated code and causal detection of code “smells”, a 1st-place finish at Microsoft’s MAIDAP Hackathon, and mentoring the next generation of applied scientists through university partnerships and internal research programs.

Before my current role, I completed my doctoral research in Computer Science at William & Mary and held research internships at Microsoft and Cisco, developing novel interpretability methods for large language models for code. My main interests are Interpretable Agentic AI, Neurosymbolic Methods, Causal AI for Software Engineering, and Quality Assurance for AI Systems. I am passionate about advancing software engineering with cutting-edge artificial intelligence and causal inference, bridging rigorous research with measurable, real-world impact.

On a personal level, I was born in Bogota (Colombia). My first language is Spanish. However, I learned English and German for professional purposes. My hobbies include biking, kayaking, hiking, and (rarely) video games. I also enjoy reading sci-fi literature. My favorite author is Isaac Asimov. Because of my Colombian heritage, I am fond of coffee in all its preparation ways.

selected publications

TSE

Toward a Theory of Causation for Interpreting Neural Code Models

David N. Palacio, Alejandro Velasco, Nathan Cooper, and 3 more authors

IEEE Transactions on Software Engineering, 2024

Abs DOI arXiv HTML Code

Neural Language Models of Code, or Neural Code Models (NCMs), are rapidly progressing from research prototypes to commercial developer tools. As such, understanding the capabilities and limitations of such models is becoming critical. However, the abilities of these models are typically measured using automated metrics that often only reveal a portion of their real-world performance. While, in general, the performance of NCMs appears promising, currently much is unknown about how such models arrive at decisions. To this end, this paper introduces docode, a post hoc interpretability method specific to NCMs that is capable of explaining model predictions. docode is based upon causal inference to enable programming language-oriented explanations. While the theoretical underpinnings of docode are extensible to exploring different model properties, we provide a concrete instantiation that aims to mitigate the impact of spurious correlations by grounding explanations of model behavior in properties of programming languages. To demonstrate the practical benefit of docode, we illustrate the insights that our framework can provide by performing a case study on two popular deep learning architectures and ten NCMs. The results of this case study illustrate that our studied NCMs are sensitive to changes in code syntax. All our NCMs, except for the BERT-like model, statistically learn to predict tokens related to blocks of code (\eg brackets, parenthesis, semicolon) with less confounding bias as compared to other programming language constructs. These insights demonstrate the potential of docode as a useful method to detect and facilitate the elimination of confounding bias in NCMs.
TOSEM

A Systematic Literature Review on the Use of Deep Learning in Software Engineering Research

Cody Watson, Nathan Cooper, David Nader Palacio, and 2 more authors

ACM Trans. Softw. Eng. Methodol., Mar 2022

Abs DOI arXiv HTML

An increasingly popular set of techniques adopted by software engineering (SE) researchers to automate development tasks are those rooted in the concept of Deep Learning (DL). The popularity of such techniques largely stems from their automated feature engineering capabilities, which aid in modeling software artifacts. However, due to the rapid pace at which DL techniques have been adopted, it is difficult to distill the current successes, failures, and opportunities of the current research landscape. In an effort to bring clarity to this cross-cutting area of work, from its modern inception to the present, this article presents a systematic literature review of research at the intersection of SE & DL. The review canvasses work appearing in the most prominent SE and DL conferences and journals and spans 128 papers across 23 unique SE tasks. We center our analysis around the components of learning, a set of principles that governs the application of machine learning techniques (ML) to a given problem domain, discussing several aspects of the surveyed work at a granular level. The end result of our analysis is a research roadmap that both delineates the foundations of DL techniques applied to SE research and highlights likely areas of fertile exploration for the future.
GECCO

Assessing Single-Objective Performance Convergence and Time Complexity for Refactoring Detection

David Nader-Palacio, Daniel Rodrı́guez-Cárdenas, and Jonatan Gomez

In Proceedings of the Genetic and Evolutionary Computation Conference Companion, Kyoto, Japan, 2018

Abs DOI HTML

The automatic detection of refactoring recommendations has been tackled in prior optimization studies involving bad code smells, semantic coherence and importance of classes; however, such studies informally addressed formalisms to standardize and replicate refactoring models. We propose to assess the refactoring detection by means of performance convergence and time complexity. Since the reported approaches are difficult to reproduce, we employ an Artificial Refactoring Generation (ARGen) as a formal and naive computational solution for the Refactoring Detection Problem. ARGen is able to detect massive refactoring’s sets in feasible areas of the search space. We used a refactoring formalization to adapt search techniques (Hill Climbing, Simulated Annealing and Hybrid Adaptive Evolutionary Algorithm) that assess the performance and complexity on three open software systems. Combinatorial techniques are limited in solving the Refactoring Detection Problem due to the relevance of developers’ criteria (human factor) when designing reconstructions. Without performance convergence and time complexity analysis, a software empirical analysis that utilizes search techniques is incomplete.