Skip to content
View kbannie's full-sized avatar
🤯
🤯

Organizations

@Backend-graduate @Dukvis @Corner-1th-Hackathon @DS-laver @jinakayoo @ezpz4 @Weather-of-Ingredients

Block or report kbannie

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.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
kbannie/README.md

header

Hi there 👋

🤖 AI Agent / RAG Engineer (Streamlit • LangChain • Pinecone)

🏆 Samsung AI Competition (Document Understanding) — 3rd / 80 teams

🔎 I build evaluation-driven LLM applications and document understanding pipelines.





Featured Work ✨

🔹 RAG-based AI Agent (Samsung Product Recommendation)

  • Streamlit-based web RAG chatbot (LangChain + Pinecone, text-embedding-3-large)
  • Added reranking (Cohere) and improved Precision 0.40 → 0.70 on 100 QA pairs
  • Guardrails via few-shot prompting + query preprocessing (dictionary/normalization)

🔹 Document Understanding (Samsung AI Competition — 3rd/80)

  • Fine-tuned DocLayout-YOLO + reading-order reconstruction
  • W&B-tracked sweeps for category-wise NMS IoU thresholds → public score +0.051 (0.371 → 0.422)
  • Layout detection improved by +0.475 F1 on multi-column documents

🔹 SPTC Prompting (Single-Pass Tree Chain-of-Thought)

  • Implemented single-pass generation → self-evaluation → selection prompting
  • ~1.5× faster inference than ToT under an SLM setup (CommonsenseQA/OpenBookQA)


Skills 🧩

Languages ✍

Python Java C C++ JavaScript HTML CSS

Tools 🔨

Streamlit LangChain Pinecone Weights & Biases Docker AWS Django MongoDB Git



Pinned Loading

  1. Machine_Learning_study_TAVE13th Machine_Learning_study_TAVE13th Public

    Jupyter Notebook