Dhruvin G.

𝗠𝗼𝘀𝘁 𝗔𝗜 𝗽𝗿𝗼𝗷𝗲𝗰𝘁𝘀 𝗻𝗲𝘃𝗲𝗿 𝗿𝗲𝗮𝗰𝗵 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻. 𝗜 𝗳𝗶𝘅 𝘁𝗵𝗮𝘁. I’m a Generative AI Engineer specializing in production-ready LLM systems - RAG pipelines, knowledge graphs, intelligent agents, document AI, and AI-powered APIs that solve real business problems, not just look good in demos. I’ve built AI systems for startups and growing teams across healthcare, hospitality, SaaS, enterprise support, and B2B platforms - systems that run reliably in production, handle real users, and deliver measurable results. 𝗥𝗲𝗮𝗹 𝗜𝗺𝗽𝗮𝗰𝘁 𝗔𝗰𝗿𝗼𝘀𝘀 𝗜𝗻𝗱𝘂𝘀𝘁𝗿𝗶𝗲𝘀 ➞ Healthcare : HIPAA-aligned clinical trial platform with patient, trial manager and support portals (AWS AppSync, Cognito, FastAPI, PostgreSQL). Hybrid agentic AI on Amazon Bedrock - the AI reads language while rules decide, trial matching runs on a knowledge graph (Apache AGE) with vector search (pgvector), and staff approve every AI reply ➞ Hospitality : WhatsApp AI assistant, guest preference engine, sentiment detection, 24/7 automated engagement ➞ SaaS & Enterprise : RAG-based internal knowledge systems, multi-agent evaluation pipelines, AI copilots ➞ B2B Platforms : Company intelligence systems, AI reasoning engines, automated data collection and scoring ➞ Content & Media : Agentic LLM pipelines for content planning, multi-agent media generation workflows 𝗪𝗵𝗮𝘁 𝗜 𝗕𝘂𝗶𝗹𝗱 RAG Systems : knowledge base AI, semantic search, document retrieval pipelines Knowledge Graphs : GraphRAG, entity and relationship modeling, graph-based matching with Neo4j and Apache AGE Multi-Agent Workflows : LangGraph, CrewAI, AutoGen - stateful agents with memory and tool use LLM Backends : FastAPI, async Python, Pydantic v2, structured outputs, tool calling MCP Integrations : custom MCP servers connecting LLMs to your data, APIs and internal tools Serverless APIs : AWS AppSync (GraphQL), Cognito authentication, role-based multi-portal access Document AI : ingestion pipelines, chunking strategies, embedding generation, vector retrieval AI Automation : n8n workflows, API integrations, business process automation Chatbots : WhatsApp, web, enterprise support - with escalation logic and personalization Data Pipelines : multi-source ingestion, scraping, cleaning, and structured storage 𝗧𝗲𝗰𝗵 𝗦𝘁𝗮𝗰𝗸 & 𝗧𝗼𝗼𝗹𝘀 AI / LLMs OpenAI, Claude (Anthropic), Gemini, Grok, Amazon Bedrock - production API integration, prompt engineering, fine-tuning Frameworks & Orchestration LangChain, LangGraph, CrewAI, AutoGen, MCP, ReAct architecture, multi-agent systems RAG & Vector Databases Pinecone, ChromaDB, FAISS, Weaviate, pgvector, hybrid search, semantic retrieval Knowledge Graphs Neo4j, Apache AGE, Cypher queries, GraphRAG Backend & Infrastructure Python, FastAPI, Asyncio, Pydantic v2, REST APIs, GraphQL, Docker, CI/CD Cloud AWS (EC2, S3, Bedrock, AppSync, Cognito), Azure (Azure ML, Azure OpenAI), GCP (VM, Vertex AI) Databases & Caching PostgreSQL, MongoDB, Redis (semantic caching) Automation & Integrations n8n, Selenium, WhatsApp Business API, webhook integrations 𝗛𝗼𝘄 𝗜 𝗪𝗼𝗿𝗸 Production first, not demos Clean architecture, maintainable code Fast build, quick iteration Direct and clear communication throughout 𝗪𝗵𝗲𝗻 𝗜’𝗺 𝗔 𝗚𝗼𝗼𝗱 𝗙𝗶𝘁 You need an AI system that actually runs in production You want RAG, knowledge graphs, agents, or LLM automation built properly You work in a regulated industry and need AI with guardrails and human oversight You are a startup or growing team going from concept to working product You want clean code you can maintain and scale after I deliver 𝗟𝗲𝘁’𝘀 𝗕𝘂𝗶𝗹𝗱 Rate: $15/hr | Available to start immediately If your project involves LLMs, RAG, knowledge graphs, agents, or AI automation - send me a message with your use case and I will tell you exactly how I would build it.