Skip to content
View Ayushhgit's full-sized avatar
🔒
Locked in
🔒
Locked in
2 followers · 2 following

Achievements

Achievement: Pull Sharkx2Achievement: YOLO

Achievements

Achievement: Pull Sharkx2Achievement: YOLO

Block or report Ayushhgit

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
Ayushhgit/README.md

MasterHead

Typing SVG

Profile Views Followers Focus


AI Animation

⚡ About Me

class AIEngineer:
    def __init__(self):
        self.name = "Ayush Raj"
        self.role = "AI/ML Engineer"
        self.language_spoken = ["en_US", "hi_IN"]
        
    def current_focus(self):
        return [
            "🔬 LLM Fine-Tuning & Quantization",
            "🏗️ RAG Architecture Design", 
            "🧠 Neural Reranking Systems",
            "⚙️ Model Optimization for Edge"
        ]
    
    def expertise(self):
        return {
            "fine_tuning": ["QLoRA", "PEFT", "LoRA"],
            "architectures": ["Mistral-7B", "Llama", "Phi"],
            "retrieval": ["FAISS", "Bi-Encoders", "Cross-Encoders"],
            "optimization": ["4-bit/8-bit Quantization", "Gradient Checkpointing"]
        }

me = AIEngineer()

🧠 Core Expertise

🤖 Generative AI

  • Parameter-Efficient Fine-Tuning (PEFT)
  • QLoRA for Mistral-7B architectures
  • Custom digital persona creation
  • Instruction tuning & alignment

🔍 Retrieval Systems

  • Advanced RAG pipeline engineering
  • Bi-Encoder retrieval mechanisms
  • FAISS vector indexing
  • Cross-Encoder neural reranking

⚙️ Model Optimization

  • 4-bit/8-bit quantization
  • Gradient checkpointing
  • Consumer GPU optimization (T4)
  • Memory-efficient training

🎯 Current Focus

  • Context-grounded generation
  • Scaling neural retrieval
  • Domain-specific LLM applications
  • Production ML systems

🛠️ Tech Arsenal

🧠 AI/ML Frameworks

🔧 LLM & Vector Tools

💻 Languages & Tools

📊 GitHub Analytics

---

🌐 Let's Connect

     


⚡ "Training models by day, optimizing gradients by night" ⚡

Pinned Loading

  1. Shortify Shortify Public

    JavaScript 1 1

  2. predictive-maintenance predictive-maintenance Public

    An end-to-end MLOps project for predicting Remaining Useful Life (RUL) of industrial equipment using the NASA C-MAPSS Turbofan Engine Degradation Dataset. This project demonstrates production-grade…

    Jupyter Notebook

  3. price-prediction-multimodal price-prediction-multimodal Public

    Forked from hari715om/price-prediction-multimodal

    This repository contains a solution for the Amazon ML Challenge 2025 where the goal is to predict product prices based on multimodal data, including product descriptions (text) and images. The proj…

    Jupyter Notebook

  4. virality-forecasting-MLOps virality-forecasting-MLOps Public

    A production-grade ML system for predicting content virality using graph neural networks, temporal modeling, and uncertainty quantification.

    Python

  5. LivingBench LivingBench Public

    A Continually Updating, Self-Auditing Benchmark for LLM Reasoning, Tool Use, and Learning Over Time

    Python

  6. DisAgent DisAgent Public

    A dynamic multi-agent system for building full-stack applications using LLM-powered agents.

    Python 1