Hugging Face's logo

Spaces:
Bhishaj
/
MerchFlow-AI
Sleeping

App Files Community
Gaurav vashistha commited on
Commit
f4ae8bb
·
1 Parent(s): 1a0a513

Docs: Complete README rewrite for production Glassmorphism UI and google-genai migration

Browse files
Files changed (1) hide show
  1. README.md +129 -39
README.md CHANGED
@@ -7,69 +7,159 @@ sdk: docker
7
  pinned: false
8
  ---
9
 
10
- # MerchFlow AI
11
 
12
- ![Python](https://img.shields.io/badge/Made%20with-Python-blue.svg)
13
- ![FastAPI](https://img.shields.io/badge/FastAPI-005571?style=flat&logo=fastapi)
14
- ![Docker](https://img.shields.io/badge/Docker-2496ED?style=flat&logo=docker&logoColor=white)
15
 
16
- MerchFlow AI is a high-performance, multi-agent orchestration system designed to automate the generation of premium e-commerce product listings. By synergizing Computer Vision, Retrieval Augmented Generation (RAG), and Large Language Models (LLMs), it transforms raw product images into SEO-optimized market-ready content.
17
 
18
- ## 🏗️ Architecture Flow
 
 
 
 
19
 
20
- The system employs a sophisticated event-driven architecture orchestrated by **FastAPI**:
21
 
22
- 1. **👁️ Visual Agent (Gemini 1.5)**
23
- * **Function**: Zero-shot image analysis.
24
- * **Process**: Extracts high-fidelity visual attributes including dominant color palettes, stylistic classifications, and granular item types.
25
- * **Engine**: Google Gemini 1.5 Flash (Vision).
 
 
 
 
 
 
 
 
 
 
 
 
 
26
 
27
- 2. **🧠 Memory Agent (Pinecone)**
28
- * **Function**: Semantic Search & RAG.
29
- * **Process**: Vectorizes visual tags to query a high-dimensional index, retrieving historically high-performing SEO keywords and market trends relevant to the product.
30
- * **Engine**: Pinecone Vector Database.
31
 
32
- 3. **✍️ Writer Agent (Llama 3)**
33
- * **Function**: Creative Synthesis.
34
- * **Process**: Fuses visual data with retrieved market intelligence to generate persuasive, conversion-focused title, description, and feature bullets.
35
- * **Engine**: Meta Llama 3 (via Groq Cloud).
36
 
37
- 4. **⚙️ Orchestrator (FastAPI)**
38
- * **Function**: Async Pipeline Management.
39
- * **Process**: Handles non-blocking agent execution, error propagation, and API lifecycle management.
 
40
 
41
- 5. **🔗 Post-Processing (n8n)**
42
- * **Function**: Automation Webhook.
43
- * **Process**: Triggers downstream workflows (database storage, Shopify API integration) via secure webhooks upon successful generation.
 
44
 
45
- ## 🚀 Complete Setup
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
46
 
47
- To run this system locally, ensure you have the following environment variables configured in your `.env` file:
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
48
 
49
  ```env
50
- GEMINI_API_KEY=your_gemini_key
51
- GROQ_API_KEY=your_groq_key
52
- PINECONE_API_KEY=your_pinecone_key
53
- N8N_WEBHOOK_URL=your_n8n_webhook_url
54
  ```
55
 
56
- ## ⚡ Quick Start
57
-
58
- ### 1. Installation
59
- Install the required dependencies using pip:
60
 
61
  ```bash
62
  pip install -r requirements.txt
63
  ```
64
 
65
- ### 2. Execution
66
- Launch the FastAPI server:
67
 
68
  ```bash
69
  python main.py
70
  ```
71
 
72
- The API will be available at `http://localhost:7860`.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
73
 
74
  ---
75
- **🛡️ Maintained by Bhishaj Technologies (UDYAM-UP-02-0108589). All Rights Reserved.**
 
 
 
 
 
 
7
  pinned: false
8
  ---
9
 
10
+ <div align="center">
11
 
12
+ # 🚀 MerchFlow AI
 
 
13
 
14
+ ### Autonomous E-Commerce Catalog Intelligence
15
 
16
+ ![Python](https://img.shields.io/badge/Python-3.10+-3776AB?style=for-the-badge&logo=python&logoColor=white)
17
+ ![FastAPI](https://img.shields.io/badge/FastAPI-005571?style=for-the-badge&logo=fastapi&logoColor=white)
18
+ ![Google Gemini](https://img.shields.io/badge/Gemini_2.5_Flash-4285F4?style=for-the-badge&logo=google&logoColor=white)
19
+ ![Pinecone](https://img.shields.io/badge/Pinecone-000000?style=for-the-badge&logo=pinecone&logoColor=white)
20
+ ![Docker](https://img.shields.io/badge/Docker-2496ED?style=for-the-badge&logo=docker&logoColor=white)
21
 
22
+ **Proprietary & Confidential — Bhishaj Technologies**
23
 
24
+ ---
25
+
26
+ *A multi-agent AI pipeline that converts raw product imagery into enterprise-grade, SEO-optimized e-commerce catalogs in seconds.*
27
+
28
+ </div>
29
+
30
+ ---
31
+
32
+ ## 🎯 Core Value Proposition
33
+
34
+ MerchFlow AI eliminates the manual bottleneck of product catalog creation. By orchestrating **Computer Vision**, **Retrieval-Augmented Generation (RAG)**, and **Large Language Models** in a seamless autonomous pipeline, it delivers production-ready product listings — from a single image upload — with zero human intervention.
35
+
36
+ ---
37
+
38
+ ## 🏗️ System Architecture
39
+
40
+ The system employs a high-performance, event-driven architecture orchestrated by **FastAPI**:
41
 
42
+ ### 👁️ Visual Analyst Agent
43
+ - **Function**: Zero-shot product image analysis
44
+ - **Process**: Extracts granular visual attributes — dominant colors, material composition, design style, branding elements, and product classification
45
+ - **Engine**: `Gemini 2.5 Flash` via the unified **Google GenAI SDK** (`google-genai`)
46
 
47
+ ### 🧠 Semantic Memory Agent
48
+ - **Function**: RAG-based keyword retrieval with intelligent fallback
49
+ - **Process**: Vectorizes visual attributes to query a high-dimensional index, retrieving historically high-converting SEO keywords and market trends. When the database has no match for a niche, the **Intelligence Fallback** system autonomously generates keywords via Gemini — ensuring **0% empty results**
50
+ - **Engine**: `Pinecone Vector DB` with `gemini-embedding-001` embeddings (768 dimensions)
51
 
52
+ ### ✍️ Writer Agent
53
+ - **Function**: High-conversion copy synthesis
54
+ - **Process**: Fuses visual intelligence with retrieved market data to generate persuasive, conversion-optimized titles, descriptions, and feature bullet points
55
+ - **Engine**: `Meta Llama 3.3 70B` (via Groq Cloud)
56
 
57
+ ### ⚙️ Pipeline Orchestrator
58
+ - **Function**: Async pipeline management & delivery
59
+ - **Process**: Handles non-blocking agent execution, error propagation, and API lifecycle management. Results are delivered instantly through the **Premium Glassmorphism UI**
60
+ - **Engine**: `FastAPI` with async/await architecture
61
 
62
+ ---
63
+
64
+ ## 🖥️ Production Interface
65
+
66
+ MerchFlow AI ships with a **Premium Glassmorphism UI** built for instant catalog generation:
67
+
68
+ - 🎨 Frosted-glass aesthetic with dynamic gradient backgrounds
69
+ - 📤 Drag-and-drop image upload with real-time processing feedback
70
+ - 📊 Structured JSON output display for visual data, SEO keywords, and generated listings
71
+ - 📱 Fully responsive design across desktop, tablet, and mobile
72
+
73
+ ---
74
+
75
+ ## 🛠️ Technology Stack
76
+
77
+ | Layer | Technology | Purpose |
78
+ |-------|-----------|---------|
79
+ | **Runtime** | Python 3.10+ | Core language |
80
+ | **Framework** | FastAPI | Async API orchestration |
81
+ | **AI SDK** | `google-genai` (Unified) | Vision & embedding inference |
82
+ | **Vision Model** | Gemini 2.5 Flash | Product image analysis |
83
+ | **Embeddings** | `gemini-embedding-001` | 768-dim vector generation |
84
+ | **Vector DB** | Pinecone (Serverless) | Semantic keyword retrieval |
85
+ | **LLM** | Llama 3.3 70B (Groq) | Copywriting synthesis |
86
+ | **UI** | Glassmorphism / Tailwind CSS | Production dashboard |
87
+ | **Deployment** | Docker / Hugging Face Spaces | Containerized hosting |
88
+
89
+ ---
90
 
91
+ ## 📋 System Updates & Technical Milestones
92
+
93
+ | Date | Milestone |
94
+ |------|-----------|
95
+ | **Feb 2026** | ✅ Full migration from deprecated `google-generativeai` to unified `google-genai` SDK |
96
+ | **Feb 2026** | ✅ Vision model upgraded to `Gemini 2.5 Flash` for industry-leading latency |
97
+ | **Feb 2026** | ✅ Pinecone vector alignment to `768 dimensions` with `gemini-embedding-001` for mathematical precision |
98
+ | **Feb 2026** | ✅ **Intelligence Fallback** system deployed — guarantees 0% empty SEO keyword results |
99
+ | **Feb 2026** | ✅ n8n webhook decoupled — pipeline relies strictly on the Glassmorphism UI for delivery |
100
+ | **Feb 2026** | ✅ Production Glassmorphism dashboard launched |
101
+
102
+ ---
103
+
104
+ ## 🚀 Quick Start
105
+
106
+ ### 1. Environment Configuration
107
+
108
+ Create a `.env` file in the project root with the following keys:
109
 
110
  ```env
111
+ GEMINI_API_KEY=your_google_genai_api_key
112
+ GROQ_API_KEY=your_groq_cloud_api_key
113
+ PINECONE_API_KEY=your_pinecone_api_key
 
114
  ```
115
 
116
+ ### 2. Installation
 
 
 
117
 
118
  ```bash
119
  pip install -r requirements.txt
120
  ```
121
 
122
+ ### 3. Launch
 
123
 
124
  ```bash
125
  python main.py
126
  ```
127
 
128
+ The production dashboard will be available at `http://localhost:7860`.
129
+
130
+ ---
131
+
132
+ ## 📁 Project Structure
133
+
134
+ ```
135
+ MerchFlow-AI/
136
+ ├── main.py # FastAPI orchestrator & pipeline logic
137
+ ├── dashboard.html # Glassmorphism production UI
138
+ ├── Dockerfile # Container deployment config
139
+ ├── requirements.txt # Python dependencies
140
+ ├── agents/
141
+ │ ├── visual_analyst.py # Gemini 2.5 Flash vision agent
142
+ │ ├── memory_agent.py # Pinecone RAG + embedding agent
143
+ │ └── writer_agent.py # Llama 3.3 copywriting agent
144
+ └── .env # Environment variables (not tracked)
145
+ ```
146
+
147
+ ---
148
+
149
+ ## ⚖️ Legal Notice
150
+
151
+ > **PROPRIETARY & CONFIDENTIAL**
152
+ >
153
+ > This software, including all source code, documentation, algorithms, and associated intellectual property, is the exclusive proprietary property of **Bhishaj Technologies (UDYAM-UP-02-0108589)**.
154
+ >
155
+ > Unauthorized copying, distribution, modification, reverse engineering, or any form of reproduction of this software — in whole or in part — is strictly prohibited and may result in legal action under applicable intellectual property laws.
156
+ >
157
+ > **© 2026 Bhishaj Technologies. All Rights Reserved.**
158
 
159
  ---
160
+
161
+ <div align="center">
162
+
163
+ **Built with precision by Bhishaj Technologies** 🇮🇳
164
+
165
+ </div>