Portfolio.ipynb Last Checkpoint: a few seconds ago (autosaved)
Epoch 42/100
Loss 0.1423
VRAM 18.4G
About Skills Projects Experience Contact

Ayush Vishwakarma



I like working with ML models, LLMs, RAG pipelines, MLOps architectures, and agent-based automation systems to build end-to-end production AI solutions.

Skills

Python SQL NLP LLMs MLOps FastAPI Docker LangChain Autogen Google ADK Agentic AI

Projects

DataPilot — Agentic AI

Multi-agent Data Science workflow automation system powered by Google ADK.

  • Google ADK
  • Python
  • Kaggle MCP
  • FireCrawl MCP
  • Prompt Engineering
  • Gemini API

Laptop Price Prediction MLOps Project on GCP

End-to-End MLOps project for predicting laptop prices using a Stacking Regressor model. It uses GCP,DVC,MLFLOW and a full ML pipeline

  • Python
  • Machine Learning
  • Gcp
  • Dvc
  • Mlflow
  • Fastapi
  • Docker

Crypto Whitepaper RAG Bot

Interact with any crypto whitepaper using Agentic RAG and Milvus Vector DB.

  • RAG
  • Web-Scraping
  • LangChain
  • Milvus
  • Gemini

MLOps Capstone

Full ML lifecycle architecture: pipelines, tracking, CI/CD & deployment.

  • Python
  • Scikit-learn
  • Dagshub
  • FastAPI
  • DVC
  • Docker
  • MLflow

Toxic Comment Analyzer

Fine-tuned DistilBERT toxicity classification for YouTube comments.

  • NLP
  • BERT
  • FastAPI
  • Streamlit

Vehicle Price Prediction

End to End Predicting the selling price of vehicles using machine learning models.

  • Python
  • pandas
  • numpy
  • matplotlib
  • XGBoost
  • FastAPI

Data Warehouse Project

Builds a structured SQL warehouse, transforming raw CRM data into query-ready analytical reporting layers.

  • SQL
  • ETL
  • Data Warehouse

Experience

Machine Learning Intern — Unified Mentor Pvt. Ltd.

Apr 2025 – May 2025 | Remote

  • Built classical ML models using sklearn + pandas.
  • Deployed FastAPI inference container system.
  • Automated CI/CD with GitHub pipelines.

Contact