Experience

Work experience

1. May 2026 — PresentCurrentJaipur, Rajasthan

   E—05

   Create Consciously AI

   AI/ML Intern

   • Contributed to TourIQ, an AI-powered tour revenue optimisation platform processing 649K+ booking records across multiple tour operators.
   • Improved booking forecast accuracy from 82% to 90% and reduced prediction MAPE from 18% to 12% using DTD, occupancy and seasonal demand features.
   • Enhanced dynamic pricing recommendations, increasing projected seat fill rates from 68% to 82% and identifying $50K+ revenue recovery opportunities across underperforming departures.
   • Worked extensively with ML pipelines, time-series forecasting, pricing optimisation and data analytics.

   • Python
   • LightGBM
   • Time-Series Forecasting
   • Dynamic Pricing
   • Pandas
   • NumPy
2. Nov 2025 — Jan 2026Remote

   E—04

   Infosys Springboard

   AI Intern

   • Built an AI-powered Python code reviewer using AST, Hugging Face (Qwen) and LangChain to detect 20+ code issues and enforce PEP8 standards.
   • Reduced linting violations from 50+ to 35 per 1,000 lines of code (~30% improvement) through automated analysis and optimisation recommendations.
   • Developed a Streamlit review platform delivering AI-generated feedback in under 2 seconds.
   • Integrated LLMs, prompt engineering and static code analysis into an end-to-end AI application.

   • Python
   • LangChain
   • Hugging Face
   • Qwen
   • AST
   • Streamlit
   • LLMs
3. Aug 2025 — Sep 2025Remote

   E—03

   Elevate Labs

   AI Intern

   • Developed a Sign Language Recognition system achieving 85.2% validation accuracy on a dataset of 10,000+ gesture images.
   • Improved model accuracy from 72.8% to 85.2% through data augmentation, feature extraction and hyperparameter optimisation.
   • Worked in an Agile team to design, test and improve ML pipelines, reducing model misclassification by ~45%.

   • Python
   • TensorFlow
   • Computer Vision
   • Deep Learning
4. Feb 2025 — Apr 2025Jodhpur, India

   E—02

   Defence Research and Development Organisation (DRDO)

   Project Trainee

   • Developed a CycleGAN-based image-to-image translation system trained on 10,000+ artwork and painting images with TensorFlow and PyTorch.
   • Improved generated image quality by reducing Fréchet Inception Distance (FID) from 58.4 to 49.6 (~15%) through preprocessing, normalisation and hyperparameter tuning.
   • Evaluated models with FID and visual quality assessment, achieving more realistic texture transfer and style preservation.

   • TensorFlow
   • PyTorch
   • CycleGAN
   • Computer Vision
5. Jan 2025 — Feb 2025Remote

   E—01

   PW (PhysicsWallah)

   Student Intern

   • Completed 3+ ML projects applying regression, classification and clustering on datasets of up to 50,000+ records.
   • Designed and evaluated models with Scikit-learn, TensorFlow and PyTorch, achieving 10–25% improvements over baseline.
   • Executed data cleaning and validation pipelines, reducing preprocessing errors by 30%.

   • Machine Learning
   • Python
   • Data Analysis