👋 Hi there! I’m Priyal Mehta

Welcome to my GitHub profile!
I'm an AI/ML Engineer up to 3 years of experience in Generative AI, Large Language Models (LLMs), NLP and Machine Learning. I have hands-on expertise in building scalable, production-grade AI systems including RAG pipelines, AI agents, and multimodal architectures.

Currently working on developing intelligent solutions such as conversational AI agents, document intelligence systems, and NLP pipelines using technologies like Python, FastAPI, LangChain, and Hugging Face Transformers. I have experience deploying models on AWS and Azure, with a strong focus on performance optimization and real-time inference.

My work includes: • Designing NLP pipelines using BERT for classification and Named Entity Recognition (NER) • Building enterprise-grade RAG chatbots with vector databases like Qdrant and Pinecone • Built a multi-tenant AI Agent using LangChain and LangGraph with tool-calling and context-aware workflows. • Developing AI voice and conversational systems using Whisper, GPT models, and Twilio • Creating multimodal AI solutions using OCR and vision-language models • Automating workflows using tools like n8n and integrating LLM-powered agents

I’m passionate about solving real-world problems using AI and continuously exploring advancements in GenAI, LLMs, and Agentic AI systems.

Open to opportunities where I can contribute to building impactful AI-driven products and scale intelligent systems.

• 🔭 Currently working on

  • AI-powered maritime systems (email intelligence, document extraction, voyage analytics)
  • LLM-based conversational agents with tool-calling & function-calling
  • RAG pipelines with vector databases for real-world business use cases

• 🧠 Core focus areas

  • Generative AI & LLMs (RAG, Agents, Prompt Engineering)
  • NLP (NER, semantic search, document intelligence)
  • Multimodal AI (OCR + Vision + LLMs)
  • Scalable backend systems for AI products

• 🌱 Currently exploring

  • Advanced RAG architectures
  • Agent orchestration & multi-tool workflows
  • Async & distributed AI systems
  • Inference optimization & cost reduction strategies

• 🤝 Open to collaborate on

  • AI / ML open-source projects
  • RAG systems, AI agents, and backend-heavy ML products

• 💬 Ask me about

  • LLMs, RAG, LangChain, FastAPI
  • NER pipelines, vector databases
  • AI system design & production deployments

• 📫 Reach me at

  • Email: [email protected]
  • LinkedIn: https://linkedin.com/in/priyal-mehta-972007233

• ⚡ Fun fact

  • I take accuracy seriously — and coffee even more seriously ☕😄

• Python
• FastAPI
• SQL, MongoDB, Redis
• Async processing & task queues

• Generative AI, LLMs, RAG
• LangChain, Flowise.ai
• Hugging Face Transformers
• Ollama, Prompt Engineering

• Named Entity Recognition (NER)
• Semantic Search
• OCR pipelines
• YOLOv8, CLIP, OpenCV
• Whisper (Speech-to-Text)
• Stable Diffusion, GANs

• Qdrant
• Pinecone

• Docker
• Git & GitHub
• AWS (S3, EC2)
• Azure
• Model deployment & inference optimization

• End-to-end AI systems (data → intelligence → decisions)
• AI agents that actually solve business problems
• Robust backend architectures for ML products
• Domain-specific AI (especially maritime & enterprise data)

⭐ If you find my work interesting, feel free to connect or collaborate!