AI Enthusiast crafting premium-grade solutions in LLM fine-tuning, NLP, CV, and Generative AI. Passionate about bringing ideas to life with precision and elegance.
Passionate about pushing the boundaries of AI and creating innovative solutions that shape the future of technology.
Mar 2025 – Present
Developing and refining an advanced grammar checker for KreativeSpace’s AI Writing Tools Suite by fine-tuning LLMs for high accuracy and natural language output. Designing and training a multilingual AI text classification system to detect AI-generated content in English, Hindi, Spanish, German, Dutch, and Italian (up to 95% accuracy), often outperforming Quillbot’s detector and approaching GPTZero.
May 2024 – Mar 2025
Built a scalable PoC of an advanced chatbot system with LangChain Agents and LLM, handling structured (SQL) and unstructured data with ~17 agents, intercommunication, RBAC, real-time ingestion, and automated source citation. Developed an OCR–LLM pipeline for Jugaad, improving invoice extraction accuracy from 15%→85% and receipts from 10%→90% across formats.
Jan 2024 – Apr 2024
Developed a generative AI system for synthetic structured data generation with Python + Gradio for cleaning, deduplication, and embedding. Built a robust PyTorch + Transformers fine-tuning framework enabling LLM adaptation across diverse datasets, significantly improving accuracy and performance, and delivered a user-friendly Gradio interface for synthesis.
Fine-tuned Qwen2.5-7B-Instruct optimized for Sanskrit language processing. Features Sanskrit to IAST transliteration, bidirectional Sanskrit ↔ English translation with context-aware preservation. Enhanced with specialized dataset, chat template format, and optimized LoRA configuration.
Native Sanskrit-English tokenizer for Qwen2.5 providing 4.5x better efficiency than byte-level tokens. Produces clean, readable tokens with 120K vocabulary trained on massive English+Sanskrit corpus. Native Hugging Face format with perfect reconstruction.
Fine-tuned Qwen2.5-7B-Instruct-1M on Sanskrit datasets using QLoRA (Axolotl), gradient checkpointing, 4-bit quantization, and cosine LR to translate Vedic Sanskrit → English.
Flux.dev model fine-tuned on different images creating various LoRAs that you can explore. Specialized LoRA adapters for different visual styles and concepts.
Transfer learning with VGG16 to identify Indian cities from images, achieving 66.3% accuracy.
Conversational agents with DialoGPT-medium: interactive chats as Tony Stark, Harry Potter, and more.
GPT-2 fine-tuned for on-demand travel itinerary generation with PyTorch + Transformers.
An infinite story generator tool with different genres and themes. Create endless narratives with customizable settings.
A CLI-based game where users can simulate real worlds and play with them. Features different models to choose from for various simulation scenarios.
A crazy 3D todo list keeper with immersive visual experience for managing your tasks.
Generate your personal quotes with AI. Share a thought, choose your vibe, and watch AI craft a unique quote card just for you. Features manual input, Twitter analysis, multiple themes, moods, and card sizes.
AI-powered platform for one-click custom chatbot creation with document ingestion (PDF/TXT), RAG via LlamaIndex, and deployment-ready workflows.