Hi, I'm Aflah, a research software engineer at Max Planck Institute for Software Systems. I specialize in backend engineering, NLP, and deep learning, with a keen interest in understanding LLMs better and leveraging NLP for social good. Currently, my research involves evaluating abilities of large language models, building benchmarks for them and decoding their learning processes such as those via memorization. I have also previously worked on hate speech normalization, hope speech detection and building recommendations for finetuning better hate speech detectors.
Open to researcher/research engineer/backend engineer rolesWorking under Dr Krishna Gummadi to explore different aspects of LLMs. Some areas we've explored/are exploring are
Working on predicting memorization behavior in LLMs by finding which strings from the training data will be memorized. Previously worked on the Pythia model suite.
Worked on multiple projects ranging from hate speech normalization to building recommendations for finetuning better hate speech detectors. Also worked on QUENCH a benchmark for advanced reasoning capabilities of LLMs with special focus on Indic context.
Worked in the Finance, Planning & Analysis Engineering division towards revamping the central hub of the department. Also built POCs based on user feedback to improve the search and access experience on the webapp. Also recieved a return offer to join full time as an Analyst.
Worked with Matthew Watson & Chen Qian towards adding support for data augmentation layers to KerasNLP a library under the Keras/TensorFlow Ecosystem which aims to build industry oriented NLP Solutions. I also contributed to several bug fixes and other utilities such as tokenizers and transformer encoder & decoder.
GPA - 9.63/10 [Dept. Rank 2 & Batch Rank 3]
Secured 95% in All India Senior School Certificate Examination
Secured 95.8% in All India Secondary School Examination
Under Review
Under Review
Under Review
Under Review
The Second Tiny Papers Track at Eleventh International Conference on Learning Representations (ICLR)
Proceedings of The 18th Conference of the European Chapter of the Association for Computational Linguistics (EACL)
Proceedings of the 15th Annual Meeting of the Forum for Information Retrieval Evaluation (FIRE)
Proceedings of the 15th Annual Meeting of the Forum for Information Retrieval Evaluation (FIRE)
Proceedings of The 40th International Conference on Machine Learning (ICML)
The First Tiny Papers Track at Eleventh International Conference on Learning Representations (ICLR)
The First Tiny Papers Track at Eleventh International Conference on Learning Representations (ICLR)
Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
PyTorch, TensorFlow, Keras, Scikit-learn, HuggingFace
Flask, FastAPI, Spring Boot
HTML, CSS, JavaScript, ReactJS, Bootstrap, Tailwind CSS, Streamlit
Python, Java, JavaScript