Skip to content
View Shivas-A54's full-sized avatar
🎯
Open to Work, Looking for Opportunities, Immediate Joiner.
🎯
Open to Work, Looking for Opportunities, Immediate Joiner.
  • Chennai, Tamil Nadu, India, Earth (Sol III), Solar System (Sol), Orion–Cygnus Arm, Milky Way Galaxy (Laniakea Zone).
  • 03:58 (UTC -12:00)

Block or report Shivas-A54

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
Shivas-A54/README.md
Typing SVG
Profile Views   Followers

🧠 About Me

╔═══════════════════════════════════════════════════════════════════╗
║  FILE: shivas.parquet                                               ║
║  FORMAT: Apache Parquet  |  ROWS: 1  |  ROW GROUPS: 1               ║
╚═══════════════════════════════════════════════════════════════════╝

SCHEMA
------
message shivas {
  required binary  name                (UTF8) = "Shivas Arulselvam";
  required binary  role                (UTF8) = "Data Analyst & Aspiring Data Engineer — AI Specialization";
  required binary  education           (UTF8) = "B.E. CSE, Anna University (2025), CGPA 7.58";
  required binary  location            (UTF8) = "Tamil Nadu, India";
  required binary  currently           (UTF8) = "Job hunting for Data Analyst & Data Engineering roles, actively learning";
  repeated binary  data_engineering    (UTF8) = ["SQL", "Python (PySpark/Pandas)", "Apache Airflow"];
  repeated binary  ai_implementation   (UTF8) = ["Predictive Maintenance", "Anomaly Detection", "Production ML"];
  repeated binary  infrastructure      (UTF8) = ["Docker", "AWS"];
  required binary  goal                (UTF8) = "Build robust, automated data architectures for adaptive AI";
}

ROW GROUP 0
-----------
  compression : SNAPPY
  encoding    : PLAIN
  num_rows    : 1

FOOTER
------
  created_by  : shivas-a54
  version     : 1.0.0

🛠️ Tech Stack & Tools

Languages & Core



Data Engineering & Warehousing

Apache Airflow Apache Spark Hadoop Pandas Data Warehousing Data Modeling


AI / ML & GenAI

Scikit-Learn XGBoost LangChain ChromaDB Ollama Gemini Groq Streamlit


Cloud & Infrastructure

AWS Docker GitHub Actions Power BI


Also in the toolkit: ETL Pipelines · Data Analysis · Jupyter Notebooks · DBMS · JDBC · Git & GitHub · OOP


🚀 Featured Projects

🔬 Project 📝 Description 🛠 Stack
💬 Prompt-to-Query (Text-to-SQL) Converts natural language into executable SQL commands via the Groq API, with instant query generation and data visualization for non-technical users Python · Streamlit · MySQL · Groq · Pandas
🔍 Stateless RAG Vector-DB-free RAG system using LLM-guided tree navigation (PageIndex API) instead of embedding-based retrieval — two-stage Gemini 2.0 Flash pipeline for section selection + cited answer generation on PDFs Python · Gemini 2.0 Flash · PageIndex API

🎓 Certifications & Learning Path

Badge Certification Issuer Date
🏭 IR 4.0 Foundation Course (Linux, Python, SQL, DBMS) TechSaksham (Microsoft & SAP Initiative) Jan 2023
📊 Big Data 101 Infosys Springboard (Wingspan) Oct 2024
📗 Data Fundamentals IBM SkillsBuild Mar 2025
🧮 SQL Mastery for Data Analytics and Business Intelligence Udemy Feb 2026
📈 Introduction to Data Science Infosys Springboard (Wingspan) Mar 2026
☁️ AWS Cloud Foundations Amazon Web Services Jul 2026

🤝 Let's Connect

LinkedIn GitHub Email


"Pipelines aren't just plumbing — they're what makes AI adaptive."

Pinned Loading

  1. Stateless_RAG Stateless_RAG Public

    A vectorless RAG pipeline that navigates PDF documents using a PageIndex tree structure and Gemini 2.0 Flash — no vector database, just LLM-guided tree search with auto-cited answers.

    Jupyter Notebook 1

  2. Prompt-to-Query Prompt-to-Query Public

    NL2SQL — turn natural language prompts into database query results

    Python

  3. sql-datawarehouse-project sql-datawarehouse-project Public

    Built a modern data warehouse using SQL, including ETL processes, data modeling, and analytics.

    TSQL 1

  4. portfolio portfolio Public

    HTML 1