OpenAlex Research Intelligence (RAG)

An interview-ready platform: hybrid + RAG search with clickable citations, four-view topic analytics, citation networks, author/institution explorers, search-log observability, an explainable pipeline journey, a guided tour, and bilingual UI — all captured end-to-end.

• Hybrid retrieval: BM25 (rank_bm25) + all-MiniLM-L6-v2 dense vectors fused with Reciprocal Rank Fusion; live facets for year, journal, type, language, and citation count.
• Citation-grounded RAG: streamed answers (SSE) where every claim carries a clickable [Wxxxx] citation; question builder + presets; extractive fallback when no LLM.
• Data Story: upload your own papers or use the sample, then watch one query travel through 7 animated pipeline stages, each with a what/effect/why explainer and a journey summary.
• Analytics: topic trends, concept co-occurrence network, concept×year heatmap, UMAP semantic cluster, journal rankings, research velocity, citation graph, author & institution explorers, and search-log analytics.
• App-like guided tour that auto-starts, spotlights each feature across pages, types out explanations, and plays themed particles; complete 中文/English toggle.

• Designing a hybrid retriever and a citation-grounded RAG pipeline that degrades gracefully to extractive mode when the LLM is offline.
• Making heavy analytics (UMAP cluster, co-occurrence graph, velocity) feel instant via in-memory TTL caching and a warm-up pipeline.
• Externalizing hundreds of UI strings into a full zh/en i18n layer while keeping technical terms in English.
• Verifying ~18 pages + the guided tour with an automated Playwright capture (viewport screenshots + a recorded walkthrough video).

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

• Detected technical signals: Python, FastAPI, React, Vite, Tailwind CSS, DuckDB, ChromaDB, BM25, sentence-transformers, RAG, llama.cpp, Recharts, Playwright
• README evidence exists and can support a fuller reviewed case study
• A public GitHub repository is linked for source traceability

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, React, Vite, Tailwind CSS, DuckDB, ChromaDB, BM25, sentence-transformers, RAG, llama.cpp, Recharts, Playwright.

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.

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

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.

• 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

• 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

• Grow the corpus beyond 100k papers and benchmark retrieval quality with qrels.
• Add multi-hop RAG and re-ranking as optional production profiles.
• Deploy the backend behind the portfolio with a reverse tunnel to a local llama.cpp.