Skip to content
View iqbalsh1's full-sized avatar
🎯
Focusing
🎯
Focusing
3 followers · 1 following
  • University of Vienna
  • Austria

Achievements

Achievement: Pull SharkAchievement: Quickdraw

Achievements

Achievement: Pull SharkAchievement: Quickdraw

Block or report iqbalsh1

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 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
iqbalsh1/README.md

IQBAL Shanda 😊

AI Engineer · Data Scientist · Data Analyst
I turn messy, real-world data into clear decisions and research ideas into working AI systems.

About Me

Working at the intersection of data analysis, machine learning, and applied AI / LLM engineering. I build end-to-end: from cleaning and analysing data, to designing reproducible ML pipelines and production-style conversational AI apps.

  • 🔭 Building Generative AI apps - RAG assistants, AI agents, and multi-provider LLM tooling
  • ✅ Focusing on grounded, reliable AI and AI Agents/Multi Agents
  • 📊 Bridging analytics and AI: EDA · statistical analysis · Power BI / Tableau dashboards · model evaluation
  • 🌱 Going deeper into Deep Learning, NLP, LLM engineering, and AI agents
  • 💬 Happy to talk Python · SQL · Data Science · Machine Learning · LLMs

Tech Stack

Languages Python SQL Bash

AI / LLM Engineering LangChain OpenAI Ollama Hugging Face ChromaDB Streamlit FastAPI

Data Analysis & BI Pandas NumPy Power BI Tableau Excel Matplotlib Seaborn

Machine Learning & Deep Learning scikit-learn PyTorch TensorFlow Keras

Tooling & Deployment Git Docker Linux

Projects

AskVista - Local RAG Policy Assistant

RAG · Privacy-first / offline · Conversational AI Ask plain-English questions about a company policy PDF and get grounded, cited answers - running 100% locally with no API keys or data leaving the machine. Embeds documents once into a persistent vector store and reuses them across sessions. Stack: LangChain · Ollama (llama3.2) · ChromaDB · Streamlit · PyPDF

Vienna City Info Bot - Context-Grounded City Assistant

Grounded LLM · Location / POI data · NLP A conversational tourism assistant that answers natural-language questions about Vienna using only a curated points-of-interest dataset - filters by category, injects the top matches as context, and never invents places. Stack: Python · OpenAI GPT-4o-mini · pandas

Weather Travel Advisor Bot - Real-Time Data + LLM

Live API integration · LLM · Real-time data Pulls live weather for any city and turns it into personalised, actionable advice (activities, what to wear, what to avoid) - a clean example of fusing a real-time API with an LLM. Stack: Python · OpenAI GPT-4o-mini · OpenWeatherMap API

More projects
  • Causal AI Benchmark Pipeline - reproducible evaluation of causal-inference algorithms · Python · scikit-learn ·
  • Data Analysis & BI Dashboard - EDA to interactive stakeholder dashboard · Pandas · Power BI / Tableau ·

Currently open to work & research opportunities - let's build something useful.

Pinned Loading

  1. weather-travel-advisor-bot weather-travel-advisor-bot Public

    AI travel advisor that reads real-time weather and suggests activities, clothing, and tips for any city.

    Python 3

  2. vienna-city-info-bot vienna-city-info-bot Public

    CLI-based AI tourism assistant for Vienna - filters a local POI dataset by category and answers natural language questions using OpenAI GPT- 4o-mini. Context-grounded, no hallucinated places.

    Python 3

  3. AskVista AskVista Public

    Local RAG chatbot that answers questions about company policy documents - built with LangChain, Ollama (llama3.2) & ChromaDB. Fully offline

    Python

  4. numerical-linear-algebra numerical-linear-algebra Public

    Numerical stability of dense LU decomposition and sparse iterative solvers - Convergence analysis and Jacobi preconditioning.

    Jupyter Notebook 2

  5. context_dependent_pairs context_dependent_pairs Public

    Context Dependent Causal Pairs

    HTML 3

  6. BrazEcomAnalytics BrazEcomAnalytics Public

    Exploratory data analysis of Brazilian e-commerce (Olist dataset) uncovering sales trends, top products, and regional insights

    Jupyter Notebook 1