Skip to content
View dasashreeya's full-sized avatar
  • College Park, MD, USA

Block or report dasashreeya

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
dasashreeya/README.md

Shreeya Dasa Lakshminath

Applied Machine Learning @ University of Maryland

LLM Systems • Machine Learning • Production-Oriented AI


About

I’m an Applied Machine Learning graduate student at UMD focused on building AI systems that go beyond experimentation.

My work sits at the intersection of:

  • machine learning
  • language models
  • system design

I’m particularly interested in how models behave in real-world settings , reliability, evaluation, and deployment.


What I Care About

  • Designing LLM systems that self-correct and adapt
  • Moving from notebooks → production ready pipelines
  • Building ML systems that are observable, testable, and scalable

Technical Signal

ML • LLMs • RAG • Retrieval Systems • Model Evaluation • Feature Engineering • MLOps


Current Direction

  • Improving reliability in LLM pipelines
  • Exploring evaluation beyond accuracy metrics
  • Building systems that integrate ML into real workflows

Elsewhere


Focused on building AI systems that hold up in the real world.

Pinned Loading

  1. agentic_rag agentic_rag Public

    Agentic RAG pipeline with self-correcting retrieval and LLM-based reasoning.

    Python

  2. crack_env crack_env Public

    Forked from ParuHere/crack_env

    YOLOv8-powered computer vision pipeline for automated crack detection in engine parts - 95% accuracy for client Vector NDT.

    Python

  3. f1-data602-project f1-data602-project Public

    Forked from ravi1-7/f1-data602-project

    Pipeline to train and predict 2025 F1 race results

    HTML

  4. np_bayes_irl np_bayes_irl Public

    Nonparametric Bayesian IRL with parallelized Gibbs sampling .

    Python 1

  5. statistical-domain-language-model statistical-domain-language-model Public

    Statistical language model fine-tuned on domain-specific corpora for specialized text generation and classification.

    Python

  6. llama_reranker llama_reranker Public

    Lightweight LoRA fine-tuning of Llama-family models as document rerankers in RAG pipelines.

    Python