Skip to content
View enescaglarr's full-sized avatar

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Block or report enescaglarr

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
enescaglarr/README.md

Hi, I'm Enes Caglar!

Data & AI/ML Engineer · M.Sc. Information Engineering, TUM (Starting Oct 2026) · B.Sc. Computer Science & Engineering, Sabanci University

enescaglar.com · LinkedIn · mail.enes.caglar@gmail.com


I like understanding systems end-to-end: from the data pipeline that feeds a model to the kernel that runs it on the edge.

Featured

  • mobilenet-edge-deployment: TensorRT FP16 and hand-written FP16 CUDA kernels for MobileNetV2, profiled on TRUBA V100 and Colab A100/L4/T4
  • wealth-management-text-to-sql: Text-to-SQL over a wealth-management warehouse: RAG grounding vs zero-shot, 30/30 vs 25/30 execution accuracy (26/26 vs 23/26 excluding docs-routed questions), with a read-only SQL guard
  • glass-box-rag: RAG customer support assistant that shows its work: retrieval scores, the exact prompt, and a repeatable eval suite (Gemini + FAISS + Streamlit)
  • inbox-triage: AI email triage for Gmail that classifies threads, labels them, drafts replies, and summarises each window (Serverless on AWS Lambda)

More

Write-ups, architecture diagrams and interactive demos for every project: enescaglar.com

Stack

Python SQL PySpark PyTorch Microsoft Fabric AWS LangChain CUDA Docker

Pinned Loading

  1. mobilenet-edge-deployment mobilenet-edge-deployment Public

    Profiling and accelerating MobileNetV2 inference — TensorRT FP16 + custom FP16 CUDA kernels, benchmarked on TRUBA V100 and Colab A100/L4/T4.

    Python

  2. wealth-management-text-to-sql wealth-management-text-to-sql Public

    Text-to-SQL over a wealth-management warehouse: RAG grounding vs zero-shot, 30/30 vs 25/30 execution accuracy, with a read-only SQL guard

    Python

  3. glass-box-rag glass-box-rag Public

    RAG customer support assistant that shows its work: retrieval scores, the exact prompt, and a repeatable eval suite. Gemini + FAISS + Streamlit.

    Python

  4. inbox-triage inbox-triage Public

    AI email triage for Gmail: classifies threads, labels them, drafts replies, and summarises each window. Serverless, free tier.

    Python