Dilara Soylu

Palo Alto, California, United States

Join Prog.AI to see contacts

Summary

🤩

Rockstar

🎓

Top School

Dilara Soylu is a PhD candidate at Stanford with nine years of software engineering experience focused on Human-Centered AI, ethics, and accessibility. She pairs strong academic training (BS from Georgia Tech, MS from Stanford, both with top GPAs) with hands-on backend contributions to high-profile open-source projects like Stanford NLP’s DSPy and the HELM evaluation framework. Her work improves robustness and data modeling—adding LM consistency checks in DSPy and refactoring MS MARCO scenarios in HELM—reflecting attention to both system reliability and evaluation rigor. Based in Palo Alto, she blends research-driven thinking with pragmatic engineering to make AI systems more transparent, usable, and responsibly designed.

9 years of coding experience

BS, Computer Science, 4.00 GPA, BS, Computer Science, 4.00 GPA at Georgia Institute of Technology

MS, Computer Science, 4.08 GPA, MS, Computer Science, 4.08 GPA at Stanford University

Github Skills (5)

computer-engineering10

python10

data-modeling9

code-optimization8

data-processing8

Programming languages (6)

JavaScriptGoPrologHTMLJupyter NotebookPython

Github contributions (5)

stanfordnlp/dspy

Aug 2024 - Feb 2025

DSPy: The framework for programming—not prompting—language models

Role in this project:

Back-end Developer

Contributions:63 reviews, 8 PRs, 56 pushes in 6 months

Contributions summary:Dilara implemented helper methods within the `dspy/primitives/program.py` and `dspy/predict/predict.py` files to control module LMs. These changes focused on structural equivalency checks for predictors and the `_assert_lm_consistency` property. The modifications involved adding methods to set, unset, and assert the consistency of Language Models (LMs) within the program's predictors, which likely contributes to the robustness of the DSPy framework.

nlpbertknowledgepredictlanguage-models

stanford-crfm/helm

Apr 2022 - Nov 2022

Holistic Evaluation of Language Models (HELM), a framework to increase the transparency of language models (https://arxiv.org/abs/2211.09110). This framework is also used to evaluate text-to-image models in HEIM (https://arxiv.org/abs/2311.04287) and vision-language models in VHELM (https://arxiv.org/abs/2410.07112).

Role in this project:

Back-end Developer

Contributions:338 commits, 1 comment in 7 months

Contributions summary:Dilara's commits primarily involve modifying and refactoring the `MSMARCOInstance` class within the `helm` repository. These changes include altering the class to be a dataclass and adjusting its attributes, indicating a focus on data model and potentially, data processing improvements. The user also introduced changes to the `msmarco_scenario.py` file. These modifications suggest the user is working on the MS MARCO scenario implementation within the Holistic Evaluation of Language Models (HELM) framework.

nlparxivabsberthelm

Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.

Request Free Trial