Skip to content
View darshan-panchal1's full-sized avatar
  • NexusLink Services India Pvt. Ltd.
  • Ahmedabad

Block or report darshan-panchal1

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 supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
darshan-panchal1/README.md

πŸ‘‹ Hi there! I'm Darshan Panchal

I'm a passionate Machine Learning Engineer with hands-on experience in developing, deploying, and optimizing AI/ML solutions. I specialize in building intelligent systems using cutting-edge technologies like Large Language Models (LLMs), deep learning, and cloud-native tools. I love turning complex problems into scalable, data-driven solutions.


πŸš€ Skills & Expertise

🧠 Machine Learning & AI

  • Deep Learning, Large Language Models (LLMs)
  • Model Fine-tuning & Prompt Engineering
  • Classification, Regression, Clustering

πŸ’» Programming & Frameworks

  • Languages: Python, SQL
  • AI/ML Frameworks: TensorFlow, PyTorch, scikit-learn
  • Data Science Libraries: NumPy, Pandas, SciPy, Seaborn, Matplotlib

πŸ› οΈ MLOps & Dev Tools

  • Git, Docker, Kubernetes, FastAPI, Jupyter Notebooks
  • Model deployment & monitoring

🌐 Additional Skills

  • Langchain, OpenAI APIs
  • AWS Bedrock, AWS Sagemaker

πŸ’Ό Work Experience

πŸ”Ή Machine Learning Engineer
Nexuslink Services India Pvt. Ltd. β€” April 2024 – Present

  • Developed and deployed ML models, improving prediction accuracy by 30%.
  • Streamlined data pipelines, reducing processing time by 40%.
  • Led EDA and feature engineering to boost model performance.
  • Built ML algorithms for classification, regression, and clustering.
  • Integrated AI systems into business operations, enhancing efficiency by 15%.
  • Monitored and fine-tuned deployed models for sustained performance.

πŸ“‚ Project Highlights

πŸ‘— Virtual Try-On App

  • Built a computer vision model enabling users to visualize outfits on themselves.
  • Used OpenPose and Detectron2 for accurate pose estimation and segmentation.
  • Trained dual U-Net models to generate realistic outputs, achieving 87% SSIM.

πŸ€– SoleyGPT Chatbot

  • Designed a Retrieval-Augmented Generation (RAG)-based chatbot for Soley Therapeutics.
  • Deployed using Docker, Kubernetes, and AWS Bedrock.
  • Integrated multiple vector databases, PostgreSQL, and MongoDB.
  • Improved chatbot response accuracy by 25% through smart context handling.

πŸ“« Let's Connect!

I'm always up for interesting collaborations and AI/ML discussions. Feel free to reach out!

LinkedIn β€’ Email


I'm enthusiastic about connecting with fellow learners, mentors, and industry professionals. Let's collaborate, learn, and grow together in the dynamic field of data science.

Pinned Loading

  1. ai-anatomy-liability-auditor ai-anatomy-liability-auditor Public

    AI-powered medical malpractice case analysis system using anatomical reasoning, legal precedent research, and multi-agent orchestration.

    Python

  2. multi-agent-collaborative-research multi-agent-collaborative-research Public

    A minimal multi-agent framework that coordinates lightweight AI agents to perform basic research tasks such as searching, summarizing, and organizing information.

    Python

  3. orbital-jurist orbital-jurist Public

    Autonomous space debris liability arbiter using multi-agent AI. Analyzes satellite conjunctions with orbital mechanics (SGP4/TLEs), applies maritime law precedents and space treaties, renders bindi…

    Python

  4. curriculum-forge curriculum-forge Public

    AI-powered educational curriculum generator with async LangGraph orchestration, Groq LLM, Qdrant vector DB, automated WCAG accessibility auditing, and Common Core standards alignment using RAG and …

    Python