Abid H.

Senior AI engineer and AI agent developer. I build AI agents, RAG chatbots, voice agents and n8n automations on OpenAI and Claude, connected to your CRM, APIs and documents, and built to keep working after launch. Top Rated Plus, 100% Job Success, 60+ Upwork jobs. Whether you're hiring an AI developer, an AI automation engineer or a chatbot developer, the work usually falls into one of these: AI AGENT DEVELOPMENT Custom AI agents that take real actions: look up records, create and update CRM entries, send emails, book appointments, and hand off to a person when they're unsure. Multi-agent systems in LangGraph and LangChain with tool calling, state, retries and human approval steps. Recent example: a deal agent for a music-industry platform that creates CRM deals through an MCP server in 6–7 seconds instead of manual entry. RAG CHATBOTS AND DOCUMENT Q&A Chatbots that answer from your own PDFs, policies and knowledge base. Chunking, hybrid search, reranking and retrieval evaluation on Pinecone, pgvector, Qdrant or ChromaDB, so answers stay accurate as the documents grow. I also fix existing RAG pipelines that give confident wrong answers. AI VOICE AGENTS AND AI RECEPTIONISTS Inbound and outbound voice agents on Vapi, Retell AI, Twilio and the OpenAI Realtime API. They answer calls, qualify leads, book appointments and log call summaries to your CRM. The same agents run on WhatsApp, Instagram, LINE, Slack and web chat. AI AUTOMATION AND CRM WORKFLOWS n8n, Make and Zapier workflows with AI steps, plus custom Python where no-code runs out. GoHighLevel, HubSpot, Follow Up Boss, Gmail and Google Sheets. Lead qualification, inbox triage, follow-ups, document and invoice processing. I also take over broken or undocumented n8n workflows. AI INTEGRATION FOR EXISTING PRODUCTS Adding LLM features to your app: FastAPI or Node.js services, OpenAI and Claude API integration, structured outputs, cost and latency tuning. Document AI for contracts, invoices and scanned forms, from OCR to clean JSON. HOW I WORK A short call to pin down the outcome, then a working prototype within days. After that we harden it: error handling, evals, logging and monitoring. Everything runs in your accounts and is documented so your team can maintain it. Six years in AI. I started in NLP research at the National Center of Artificial Intelligence, then worked as an AI engineer on multi-agent systems at Patchbay and on real estate automation at @realty. Industries: real estate and proptech, music, hospitality, SaaS and customer support. Stack: Python, FastAPI, Node.js, TypeScript, LangGraph, LangChain, LlamaIndex, OpenAI API, Claude API, Gemini, MCP, Pinecone, pgvector, Qdrant, n8n, Make, Zapier, GoHighLevel, PostgreSQL, Supabase, Docker, GCP, AWS. Send an invite with what you want built or fixed, and I'll tell you quickly whether I'm the right fit.