Skip to content
View TubaSid's full-sized avatar
👽
Focusing
👽
Focusing

Block or report TubaSid

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
TubaSid/README.md

✨ Tuba Siddiqui

Building intelligent systems, one chai at a time ☕

typing


🧩 Who I Am

Hi, I’m Tuba Siddiqui, people usually call me TeeSid, or just Tee. I am a Data Scientist who’s gradually moving from models to systems.
My world currently revolves around:

  • 🧠 Designing AI workflows that connect LLMs, APIs and logic
  • ⚙️ Exploring solution architecture for scalable AI products
  • 🪄 Automating repetitive intelligence through LangChain, n8n and FastAPI

Graduated 🎓 from Sapienza University of Rome with a Master’s in Data Science,
I’ve worked at Fastweb (Italy) and HBL (Pakistan) building data tools, dashboards and automation prototypes.


🚀 What I’m Brewing (A Few of Them)

🤌 Project Essence
❤️ Khayal An AI Journaling & Reflection Assistant: A conversational agent designed for daily check-ins, emotional reflection, and habit tracking, integrated with WhatsApp and backed by LLM tools, memory, and automation flows.
👀 Feedback Collector AI Feedback Collector (n8n + Gemini), which is an end-to-end automated system that captures user feedback, cleans it, tags themes, and generates summaries using LLM-powered workflows. 🔗 FeedbackCollector
🧠 Semantic Segmentation of Key Fetal Brain Structures in Trans-Thalamic Ultrasound Images Using Deep Learning (Master Thesis) Deep-learning segmentation with Attention U-Net + Grad-CAM to assist anomaly screening
🚦 VisionRoad (Bachelor Thesis) YOLOv3 + Raspberry Pi for adaptive traffic light control 🔗VisionRoad
🎙️ Voice Recognition System MATLAB-based DSP pipeline for speaker ID 🔗VoiceRecognitionSystem
🧩 Address Data Cleaning Tool Streamlit + RapidFuzz for fuzzy matching and postal validation

⚡ Tech Stack

Python · R · PyTorch · TensorFlow · LangChain · FastAPI · Streamlit
Docker · AWS · PowerBI · OpenCV · Pinecone · Qdrant


🎯 My 2025 Focus

  • Architect small-scale AI automation systems that actually deploy
  • Write tech + thought pieces on system design for young builders
  • Blend data science with architecture thinking
  • Stay curious enough to break my own code (and then fix it better)

🧭 Motto

AI is an orchestra and data is its melody, un’armonia d’immaginazione.


🌐 Connect with Me

📫 tubaasid@gmail.com
🌍 Portfolio


☕ Sipping chai, brewing logic, con un tocco d’intelligenza.

Pinned Loading

  1. DataPath DataPath Public

    DataPath of Single Cycle Core on Scala(Chisel) based on RISC-V ISA

    Scala

  2. Mastermind Mastermind Public

    It was a mind game written in C++. Program used to create a random sequence of 8 colours and you had to guess the right sequence in 3 tries.

    C++ 1

  3. Voice-Recognition Voice-Recognition Public

    Voice recognition is the process of identifying the speaker of the audio provided in the input by matching it with the audios stored in the database. This can be used in security purposes, which me…

    MATLAB 4 3

  4. PiplinedDataPath PiplinedDataPath Public

    Piplined DataPath

    Scala

  5. Mazar-e-Quaid Mazar-e-Quaid Public

    It is a structure of Tomb of Quaid aka Mazar-e-Quaid, drawn on C++ using graphics.h library in computer graphics.

    C++ 1

  6. Smart-Traffic-Management-System Smart-Traffic-Management-System Public

    There has been a population increase which has consequently led to traffic congestion in the city of Karachi. Making a smart traffic management system that makes use of video and picture data of th…

    Python 9