{"id":17717124,"url":"https://github.com/tilotech/identity-rag-customer-insights-chatbot","last_synced_at":"2025-04-09T05:09:53.299Z","repository":{"id":259218797,"uuid":"876647724","full_name":"tilotech/identity-rag-customer-insights-chatbot","owner":"tilotech","description":"Connect to your customer data using any LLM and gain actionable insights. IdentityRAG creates a single comprehensive customer 360 view (golden record) by unifying, consolidating, disambiguating and deduplicating data across multiple sources through identity resolution.","archived":false,"fork":false,"pushed_at":"2024-11-21T14:49:56.000Z","size":3458,"stargazers_count":215,"open_issues_count":0,"forks_count":18,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-03-31T05:45:14.539Z","etag":null,"topics":["agentic-rag","identity-resolution","langchain"],"latest_commit_sha":null,"homepage":"https://identityrag.com","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/tilotech.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-10-22T10:25:07.000Z","updated_at":"2025-03-31T04:56:38.000Z","dependencies_parsed_at":"2024-10-23T16:16:40.812Z","dependency_job_id":"2ba93fe5-bd44-44ef-af48-e8ece9fb62c2","html_url":"https://github.com/tilotech/identity-rag-customer-insights-chatbot","commit_stats":null,"previous_names":["tilotech/identity-rag-customer-insights-chatbot"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tilotech%2Fidentity-rag-customer-insights-chatbot","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tilotech%2Fidentity-rag-customer-insights-chatbot/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tilotech%2Fidentity-rag-customer-insights-chatbot/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tilotech%2Fidentity-rag-customer-insights-chatbot/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/tilotech","download_url":"https://codeload.github.com/tilotech/identity-rag-customer-insights-chatbot/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247980837,"owners_count":21027808,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agentic-rag","identity-resolution","langchain"],"created_at":"2024-10-25T14:11:11.590Z","updated_at":"2025-04-09T05:09:53.272Z","avatar_url":"https://github.com/tilotech.png","language":"Python","funding_links":[],"categories":["🛎️ Customer Ops","Python","A01_文本生成_文本对话"],"sub_categories":["🟩 Development Tools 🛠️","大语言对话模型及数据"],"readme":"\u003ch1 align=\"center\"\u003e\nIdentityRAG\u003cbr/\u003eCustomer Insights Chatbot\n\u003c/h1\u003e\n\n\u003ch3 align=\"center\"\u003e\nConnect to your customer data using any LLM and gain actionable insights.\n\u003c/h3\u003e\n\n\u003cp align=\"center\"\u003e\n| \u003ca href=\"https://identityrag.com\"\u003e\u003cb\u003eIdentityRAG\u003c/b\u003e\u003c/a\u003e | \u003ca href=\"https://identity-rag.streamlit.app\"\u003e\u003cb\u003eLive Demo\u003c/b\u003e\u003c/a\u003e | \u003ca href=\"https://linkedin.com/company/tilores\"\u003e\u003cb\u003eLinkedIn\u003c/b\u003e\u003c/a\u003e |\n\u003c/p\u003e\n\nhttps://github.com/user-attachments/assets/b6639333-54ae-4d20-a585-c8cf604036d6\n\n## What is IdentityRAG?\n\nIdentityRAG is a retrieval-augmented generation system that integrates identity\nresolution capabilities to provide accurate, context-aware responses about\nspecific customers. It retrieves unified customer data across disparate sources\nto create a comprehensive golden record before generating LLM responses,\nensuring answers are based on an accurate, deduplicated view of the customer.\n\n| LangChain Integration                          |\n|------------------------------------------------|\n| ![IdentityRAG](/docs/assets/identityRAG.gif)   |\n\n## Key Features\n* \u003cb\u003eUnify\u003c/b\u003e - bring data together from various sources.\n* \u003cb\u003eSearch\u003c/b\u003e - find and retrieve all relevant customer data with fuzzy matching.\n* \u003cb\u003eConsolidate\u003c/b\u003e - combine it meaningfully by creating a golden record.\n* \u003cb\u003eDisambiguate\u003c/b\u003e - resolve conflicts/unclear matches.\n* \u003cb\u003eDeduplicate\u003c/b\u003e - remove redundancies where repeated with no extra value.\n\n| Multiple Customer Data Sources  / Knowledge Bases |\n|---------------------------------------------------|\n| ![IdentityRAG](/docs/assets/multipleSources.gif)  |\n\n## Live Demo\nIf you don't want to use your own LLM keys then give it a try on the following live demo.\n\n| Live Demo |\n|--|\n|[![Live Demo](docs/assets/liveDemo.png)](https://identity-rag.streamlit.app)|\n\n## Getting Started\n\n1. Clone this repository\n2. Install dependencies: `pip install -r requirements.txt`\n3. Set up your environment variables (see Configuration section)\n4. Run the demo server: `chainlit run chat.py -w`\n5. Open [http://localhost:8000](http://localhost:8000) in your browser\n6. Try asking \"search for Sophie Muller\"\n\n## Configuration\n\n### 1. Configure Customer Data Access:\n```bash\nexport TILORES_API_URL='https://8edvhd7rqb.execute-api.eu-central-1.amazonaws.com'\nexport TILORES_TOKEN_URL='https://saas-umgegwho-tilores.auth.eu-central-1.amazoncognito.com/oauth2/token'\nexport TILORES_CLIENT_ID='3l3i0ifjurnr58u4lgf0eaeqa3'\nexport TILORES_CLIENT_SECRET='1c0g3v0u7pf1bvb7v65pauqt6s0h3vkkcf9u232u92ov3lm4aun2'\n```\n\n### 2. Configure LLM Access:\n#### Option 1: ChatGPT using OpenAI keys\n```bash\nexport OPENAI_API_KEY='your openAI key'\nexport OPENAI_MODEL_NAME='gpt-4o-mini'\n```\n\n#### Option 2: Anthropic Claude using AWS Bedrock keys\n```bash\nexport LLM_PROVIDER='Bedrock'\nexport BEDROCK_CREDENTIALS_PROFILE_NAME='your configured AWS profile name with bedrock access'\nexport BEDROCK_REGION='us-east-1'\nexport BEDROCK_MODEL_ID='anthropic.claude-3-5-sonnet-20240620-v1:0'\n```\n\u003e [!IMPORTANT]  \n\u003e The aws profile needs to have access to the model with action\n`InvokeModelWithResponseStream`. Also make sure the model is enabled in bedrock\nconsole and in the correct region.\n\n## Using Your Own Data\n\nTo use your own data you will need to create a Tilores instance and get your free Tilores API credentials,\nHere's how to do that:\n* Visit [app.tilores.io](https://app.tilores.io) and sign up for free.\n* Click on \"**_Switch to Instance View_**\" on the bottom right.\n* Select \"**_Upload Data File_**\" option and proceed. It is recommended to use csv file format.\n* If the file has well named headers the matching will be automatically configured and you can proceed with the instance\n  creation without any further changes. The deployment will take around 3-5 minutes.\n* Once the deployment is done, navigate to \"**_Manage Instance_**\" -\u003e \"**_Integration_**\" -\u003e \"**_GraphQL API_**\"\n* The first URL is the `TILORES_GRAPHQL_API`, and the second is `TILORES_TOKEN_URL` you will need to export these two\n  values as shown in Configuration section.\n* Then click `CREATE NEW CREDENTIALS` and store both values. Then export each one into its corresponding environment\n  value `TILORES_CLIENT_ID` and `TILORES_CLIENT_SECRET`.\n* Now run `chainlit run chat.py -w` and ask to search for one of the records in your data.\n\n## PDF Link Lookup Tool\n\nIf you want to test the automatic lookup from the PDFs, you also must have the poppler-utils installed:\n```\nsudo apt-get install poppler-utils","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftilotech%2Fidentity-rag-customer-insights-chatbot","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftilotech%2Fidentity-rag-customer-insights-chatbot","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftilotech%2Fidentity-rag-customer-insights-chatbot/lists"}