N

Nirmal Solanki

About

Detail

Madhya Pradesh, India

Timeline

work
Job
school
Education

Résumé

Jobs verified_user 0% verified
  • Passionbits
    Machine Learning Engineer
    Passionbits
    May 2025 - Current (1 year 3 months)
    Spearheading the development of a recommended architectural structure for videos to enhance analysis efficiency.Automated the analysis of market trends by harnessing data from Reddit, Facebook, and Instagram.Maintaining the backend of the whole pipeline: Database Management, Celery, caching, and APIs.
  • B
    Research Intern
    Boltzmann Labs
    Jan 2025 - Apr 2025 (4 months)
    Developed and implemented machine learning-driven force fields for precise molecular dynamics simulations of protein-ligand interactions.Analyzed datasets SPICE, QM9, ANI1, and MD, and benchmarked classical AMBER and CHARMM force fields for performance improvements.
  • T
    Machine Learning Engineer
    TRUEFAN
    Aug 2024 - Sep 2024 (2 months)
    Optimized lip-sync modules by improving facial landmark detection and facial alignment using InsightFace, enhancing lip-sync accuracy.Implemented and deployed machine learning solutions using Docker containers, and developed RESTful APIs for efficient model integration and deployment.
Education verified_user 0% verified
  • Indian Institute of Technology, Bombay
    B.Tech/B.E.: Engineering Physics
    Indian Institute of Technology, Bombay
    Jan 2020 - May 2024 (4 years 5 months)
    Projects and Research:Image Captioning with Cross-Attention for enhanced visual-language alignmentFacial Detection Bias Mitigation using Semi-Supervised VAE (SSVAE)Image Segmentation with U-Net for high-precision visual mappingLLM Training & Fine-tuning using Multi-Head Attention (MHA)
Projects (professional or personal) verified_user 0% verified
  • E
    Email Reply System
    Developed a scalable, automated email reply system for film equipment rental using LLM-based classification, RAG pipeline, and SQLAlchemy ORM Implemented context-aware response generation using LangChain, OpenAl LLMs, and a structured SQLitebacked inventory model for dynamic equipment availability handling.
  • F
    Face Segmentation Application
    Deployed Streamlit app integrating BiSeNet and Dlib to perform precise facial segmentation, mask refinement using OpenCV, contour extraction, GrabCut, and morphological operations for smooth results, optimized pipeline with rembg background subtraction, and edge enhancement for high-quality segmentation outputs
  • W
    Web Scraping and Question-Answer system
    Developed a web-based content extraction tool that integrates the Firecrawl API to scrape multiple URLs, enabling users to preview extracted data and interact through questions using Gemini's language model, and built a full-stack application using Node.js, Express, and Tailwind CSS that fetches, stores, and analyzes webpage content