Enter Instagram Reel Link & Select Product Category.
Pipeline Starts with Realtime Status Updates.
Amazon Listing Generated via Instagram Reel.
Automatic 1:1 Conversion from 16:9 with Cropping Feature.

VidChemy

Converting short-form promotional videos into marketplace-ready, SEO-optimized e-commerce listings - in minutes.

Inspiration

Small creators and influencers drive over 40% of online product discovery through short-form videos on Instagram Reels, TikTok, and YouTube Shorts.

Yet, over 70% of them struggle to convert that viral content into structured e-commerce listings efficiently.

Sellers spend 6–10 hours per product, manually:

Extracting images
Editing visuals
Analyzing competitors
Writing SEO content

They typically use 4–6 different tools, making workflows fragmented and slow - delaying product launches by up to 60% and reducing their ability to monetize viral content..

We built VidChemy to change that: a single AI-powered pipeline that automates the entire process, helping sellers go from video → live product page 10× faster.

Process Flow

VidChemy Process Flow Diagram
A high-level view of how a video is ingested, analyzed, and transformed into a final e-commerce listing.

What It Does

VidChemy transforms short-form promotional videos into fully optimized e-commerce product listings within minutes.

It:

Analyzes video, audio, and post metadata for context
Extracts the best product frame automatically
Cleans and enhances it into a 1:1 product image with background removal
Generates SEO-optimized titles, descriptions, and keywords
Runs competitor and sentiment analysis for ranking optimization

Output Quality

Vidchemy's AI-generated listings score 92/100 on SEO and 85/100 on Total Listing Score - with full marks on title optimization, keyword optimization, and bullet points, validating the effectiveness of the end-to-end pipeline.

How We Built It

We built VidChemy on a robust AI-first pipeline running fully on AWS.

VidChemy System Architecture Diagram
VidChemy's scalable, worker-based cloud architecture.

Tech Stack

Frontend: React + TypeScript + Vite + Tailwind CSS
Backend: Node.js + Express + MongoDB
Task Management: BullMQ for distributed worker pipelines
Media: FFmpeg for frame extraction & media optimization
AI & ML: Amazon Bedrock for end-to-end AI pipeline, using Nova Pro for SEO-optimized listing generation, Bedrock Video Analysis for frame/scene understanding, Amazon Transcribe for audio transcription, and Titan Image Generator for professional background removal and product image enhancement
SEO Engine: Competitor + review sentiment–based optimization
DevOps: Dockerized stack, orchestrated via Docker Compose
Deployment: AWS EC2 + Nginx reverse proxy + SSL (DuckDNS)

AWS-Powered Architecture Enhancing AI-Driven User Experience

Ingestion & Safety (AWS Bedrock Video Analysis): Automatically filters incoming Instagram Reels for relevance and safety, keeping the processing pipeline clean.
Precision Extraction (AWS Transcribe & Bedrock OCR): Combines audio transcripts with on-screen text recognition to flawlessly identify product names and features.
Visual Enhancement (AWS Bedrock Titan): Automatically removes backgrounds to instantly generate professional, manual-effort-free product imagery.
Listing Optimization (AWS Bedrock Nova Pro): Leverages competitor and sentiment data to auto-generate A10-optimized listing copy that drives engagement.
The Bottom Line (User Value): A fully automated engine that turns raw social video into high-converting, professional product listings in seconds.

Cost Overview

Component	MVP (Monthly)	Production (10k videos/mo)
Compute: AWS EC2	$0.00 (Credits)	~$120.00
Message Queue: Redis	$0.00 (Local)	~$35.00
Database: MongoDB	$0.00 (Atlas M0)	~$60.00
Storage: AWS S3	$0.00 (<5GB)	~$15.00
CDN: AWS CloudFront	$0.00 (<1TB)	~$45.00
RapidAPI Instagram	$0.00 (100 req/mo)	$20.00
HasData Amazon	$0.00 (1k credits)	$49.00
AWS Bedrock & Titan	$0.00 (Credits)	~$120.00
Total	$0.00 / month	~$464.00 / month

~$0.04 per processed video at production scale.

Challenges We Faced

Combining multiple AI APIs (Perplexity, Bedrock, scraping services) into one continuous workflow
Building clear error reporting for long async pipelines
Handling concurrent video and image processing without blocking workers
Maintaining user responsiveness despite heavy background tasks

Accomplishments

Built a production-style AI pipeline that converts social videos directly into marketplace-ready listings.
Deployed a reliable, distributed container infrastructure on AWS
Automated image cleanup, extraction, and SEO generation
Achieved a 10× faster listing workflow for sellers and creators