{"slug": "my-journey-building-ai-agents-rag-systems-and-ai-powered-applications", "title": "My Journey Building AI Agents, RAG Systems, and AI-Powered Applications", "summary": "A developer has built a portfolio of AI agent and RAG applications, including a Deep Research Agent that runs multi-stage research workflows with FastAPI, LangGraph, Tavily and ChromaDB, and an Autonomous Financial Research Agent using a ReAct-style workflow over SEC EDGAR filings, earnings transcripts and financial data. The projects also include a dental appointment bot integrating AI agents with PostgreSQL, Redis, Celery, Stripe and Google Calendar, and NexusBase, an enterprise RAG architecture with query routing, retrieval evaluation and pgvector.", "body_md": "I have been building AI applications to understand what happens when an LLM has to do more than just answer a prompt.\n\nMy projects cover research, agentic workflows, RAG, memory, automation, document processing, financial analysis, and AI-powered applications.\n\nHere are some of the projects I have built and what each one taught me.\n\n**Status:** Live\n\n**GitHub:** [Research-AI-Agent](https://github.com/unnita1235-code/Research-AI-Agent)\n\n**Live Demo:** [Research AI Agent](https://research-ai-agent-28jy.vercel.app/)\n\nThe Deep Research Agent takes a research topic and turns it into a structured research report.\n\nInstead of making one LLM call, the application goes through multiple research stages. It generates search queries, collects information, extracts facts, identifies gaps, performs follow-up searches, and then uses the collected information to create the final report.\n\nI built it with FastAPI and LangGraph, with Tavily for web research and ChromaDB for retrieval.\n\nThe project also supports report export, follow-up questions, summaries, counterarguments, translation, job history, Docker deployment, and an Ollama-based local mode.\n\n**What makes this project different:**\n\nThe focus is on the complete research workflow rather than simply putting a chatbot interface on top of an LLM.\n\n**GitHub:** [Autonomous-Financial-Research-Agent](https://github.com/unnita1235-code/Autonomous-Financial-Research-Agent)\n\n**Live Demo:** [Autonomous Financial Research Agent](https://autonomous-financial-research-agent.vercel.app/)\n\nThis project focuses on autonomous financial research.\n\nThe agent uses a ReAct-style workflow to decide what information it needs, select tools, process the results, and continue the research.\n\nIt can work with SEC EDGAR filings, earnings transcripts, financial data, news, sentiment, peer comparisons, fact checking, and calculations.\n\nI also built three types of memory:\n\nThe project also includes conflict resolution, PII redaction, prompt-injection protection, rate limiting, and an evaluation framework.\n\n**What makes this project different:**\n\nIt combines agent reasoning, multiple research tools, memory, financial data, and evaluation instead of treating research as a single search-and-answer task.\n\n**GitHub:** [Autonomous-Dental-Appointment-Bot](https://github.com/unnita1235-code/Autonomous-Dental-Appointment-Bot)\n\n**Live Demo:** [Autonomous Dental Appointment Bot](https://autonomous-dental-appointment-bot.vercel.app/)\n\nThis project applies AI to appointment automation.\n\nPatients can book, reschedule, and cancel appointments through web, SMS, WhatsApp, and voice interfaces.\n\nThe system connects the AI agent with application services, PostgreSQL, Redis, Celery, Stripe, and Google Calendar.\n\nI also worked on practical problems such as appointment slot locking, payment webhooks, duplicate event handling, logging, health checks, and error handling.\n\n**What makes this project different:**\n\nThe AI is connected to application workflows and business rules instead of functioning as an isolated conversation interface.\n\n**Status:** Backend functional / Frontend being redeployed\n\n**GitHub:** [NexusBase](https://github.com/unnita1235-code/NexusBase)\n\nNexusBase is an enterprise RAG architecture project.\n\nIt uses Next.js, FastAPI, LangGraph, PostgreSQL, and pgvector.\n\nThe system routes queries through different paths depending on what the user is asking. Retrieval-based questions can go through vector search and response evaluation before producing the final response.\n\nThe entire application is containerized with Docker.\n\n**What makes this project different:**\n\nThe project focuses on the architecture behind an enterprise RAG system, including routing, retrieval, evaluation, and structured responses.\n\n**GitHub:** [MedComply](https://github.com/unnita1235-code/medcomply)\n\n**Live Demo:** [MedComply](https://medcomply.vercel.app/)\n\nMedComply is a medical compliance SaaS project.\n\nIt is structured as a monorepo with a Next.js frontend, FastAPI backend, and Supabase database migrations.\n\nThe system includes organizations, users, documents, authentication, role-based access control, document processing, and AI-assisted analysis.\n\nThe repository also includes Docker configuration and GitHub Actions for automated checks.\n\n**What makes this project different:**\n\nThe project combines AI functionality with application-level access control, document management, and compliance-oriented workflows.\n\n**GitHub:** [Hybrid-Financial](https://github.com/unnita1235-code/Hybrid-Financial)\n\n**Live Demo:** [Aequitas FI](https://aequitas-web-phi.vercel.app/)\n\nAequitas FI combines structured financial data with RAG.\n\nInstead of treating every question as a document-search problem, the system separates structured SQL analysis from document retrieval.\n\nLangGraph coordinates the workflow, while PostgreSQL and pgvector provide the data layer.\n\nThe project also includes temporal comparison, PII redaction, audit logging, human feedback, and automated testing.\n\n**What makes this project different:**\n\nIt combines SQL and RAG in the same AI workflow so that structured data and retrieved documents can contribute to the final answer.\n\n**Status:** Architecture demonstration\n\n**GitHub:** [Context Synthesizer](https://github.com/unnita1235-code/enterprise-thread)\n\n**Live Demo:** [Context Synthesizer](https://enterprise-thread.vercel.app/)\n\nContext Synthesizer is an enterprise RAG architecture demonstration.\n\nThe project shows how information from systems such as Slack, Jira, Google Drive, and Notion could be organized into a unified retrieval workflow.\n\nThe interface demonstrates hybrid search, semantic retrieval, reranking, entity relationships, evaluation metrics, and observability.\n\nThis project is intentionally presented as an architecture demonstration.\n\nThe live external connectors, vector database, embedding pipeline, backend retrieval engine, authentication, and production LLM inference are not currently implemented.\n\n**What makes this project different:**\n\nThe focus is on visualizing and explaining the architecture required for enterprise knowledge retrieval rather than claiming a demo interface is already a complete production system.\n\nThe biggest thing I learned is that building an AI application is not just about connecting an LLM to a prompt.\n\nThe difficult parts are usually around the model.\n\nYou have to decide:\n\nThat is why my projects gradually moved from simple AI interfaces toward systems involving tools, retrieval, memory, evaluation, security, and application logic.\n\nI am still improving these systems, but building them has given me practical experience with different ways of designing AI applications.\n\nMy goal is not just to make an LLM generate an answer.\n\nI want to build systems around AI that can actually perform useful work.\n\n**AI disclosure:** This article was prepared with AI assistance for drafting and editing. The project information and technical details are based on my own projects and repositories, and I reviewed the content before publishing.", "url": "https://wpnews.pro/news/my-journey-building-ai-agents-rag-systems-and-ai-powered-applications", "canonical_source": "https://dev.to/unni_ta_1bf8d143dac1c4a0/my-journey-building-ai-agents-rag-systems-and-ai-powered-applications-4424", "published_at": "2026-09-29 21:52:55+00:00", "updated_at": "2026-09-29 22:16:45.472291+00:00", "lang": "en", "topics": ["ai-agents", "large-language-models", "generative-ai", "ai-tools", "mlops"], "entities": ["FastAPI", "LangGraph", "Tavily", "ChromaDB", "PostgreSQL", "pgvector", "Redis", "Stripe"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/my-journey-building-ai-agents-rag-systems-and-ai-powered-applications", "markdown": "https://wpnews.pro/news/my-journey-building-ai-agents-rag-systems-and-ai-powered-applications.md", "text": "https://wpnews.pro/news/my-journey-building-ai-agents-rag-systems-and-ai-powered-applications.txt", "jsonld": "https://wpnews.pro/news/my-journey-building-ai-agents-rag-systems-and-ai-powered-applications.jsonld"}}