Justin

Command Palette

Search for a command to run...

Back to project archive
Prototype2026ai-data

Amazon Review Intelligence

A review-centered product search and recommendation data product.

Amazon Review Intelligence combines Amazon Reviews 2023 data, BM25, vector search, reranking, review summarization, sentiment and aspect analysis, similar-product search, recommendation APIs, and a dashboard experience.

Live DemoView GitHubDemo RecordingREADME

Verified Demo Recording 1

Copied by the portfolio quality pass from existing project demo media.

12 screenshots1 demo videos

Media overview

Browse the project screenshots and recorded walkthroughs.

Project links and demo readiness

live

Live Demo

Open the verified live demo when available, or use the internal run guide for manual demo preparation.

github

View GitHub

Open the public repository when available, or review the source-access status captured by this case study.

video

Demo Recording

Open the demo recording when available, or use this section as the recording checklist for the project.

documentation

README

Open the project README when available, or use this case study as the documentation baseline.

Project overview

Amazon Review Intelligence is a review-centered product search and recommendation data product. It combines Amazon Reviews 2023 category data, BM25, vector search, reranking, review summarization, sentiment and aspect mining, similar-product search, recommendation APIs, and a dashboard experience. The project turns review data into a searchable, comparable, and evaluable product experience. The backend uses Python, DuckDB, and FastAPI for data and API layers, while the frontend uses Next.js for product search, review summaries, recommendations, comparison, embedding clusters, pipeline visibility, and evaluation screens. It also includes pytest and Playwright coverage so data processing, API behavior, and main frontend workflows can be checked repeatedly. The portfolio card should communicate that this is not just a dashboard. It is a full data product exercise spanning ingestion, indexing, retrieval, recommendation, evaluation, and an operator-facing interface.

My role

Independent Developer

Problem

Product pages contain many reviews, but shoppers cannot easily turn raw review volume into clear tradeoffs, sentiment patterns, comparable products, or recommendation signals.

Solution

Built a review-centered data product with DuckDB for product and review data, FastAPI for search and recommendation endpoints, and a Next.js dashboard for hybrid search, product summaries, sentiment analytics, recommendations, and evaluation views.

Highlights

  • **Hybrid Search** — BM25 lexical + sentence-transformer vector search with configurable alpha blending
  • **Review Intelligence** — per-product summaries, pros/cons extraction, sentiment distribution
  • **Recommendation** — popularity, content-based, item-item, and cold-start strategies
  • **REST API** — FastAPI with 8 endpoints, OpenAPI docs at `/docs`

Engineering challenges

  • Balanced BM25 keyword precision with sentence-transformer semantic recall through configurable hybrid ranking.
  • Kept the portfolio presentation public-safe by shipping code, docs, synthetic samples, screenshots, and aggregate metrics instead of redistributing raw Amazon review dumps.

Target users

  • Portfolio reviewers and interviewers
  • Technical readers who need a quick view of purpose, stack, and maturity

Technical highlights

  • Detected technical signals: Python, FastAPI, Next.js, TypeScript, DuckDB, BM25, Sentence Transformers, scikit-learn, Playwright
  • README evidence exists and can support a fuller reviewed case study
  • A public GitHub repository is linked for source traceability

Architecture

This case study is generated from the portfolio catalog pipeline using README, Git metadata, package/build configuration, and media signals. The final architecture narrative still needs source-level review. Current detected technology signals include: Python, FastAPI, Next.js, TypeScript, DuckDB, BM25, Sentence Transformers, scikit-learn, Playwright.

Data flow

A public data-flow narrative is not fully reviewed yet. If the project includes data processing, AI pipelines, or backend APIs, the next pass should document input, processing, storage, and UI/output flow end to end.

Project structure

amazon-review-intelligence/
  README.md              # project documentation, when available
  source files           # implementation reviewed by local audit
  package/build config   # detected capability signals

Setup / Run guide

This project does not expose a verified runnable web command yet. Review the README/source tree and add exact install, run, test, and build commands before interview use.
No verified build command was detected. Treat the current portfolio page as a case-study placeholder until build steps are reviewed.

Future improvements

  • Complete the production-quality README, screenshots, and demo recording
  • Add architecture diagrams, data-flow notes, and key technical decisions
  • Verify build/test status and update the portfolio release report

Interview notes

  • State the current maturity and demonstrable scope first
  • Focus on verified stack, source structure, and completed behavior
  • Do not claim unverified deployment, video, or test coverage as finished

Next steps

  • Add a public demo link and a fuller walkthrough video.

Related projects