File size: 1,060 Bytes
cb4d79c
6bf9172
1aebf33
 
 
6bf9172
 
cb4d79c
06522fc
cb4d79c
 
1aebf33
cb4d79c
6bf9172
 
cb4d79c
1aebf33
 
 
 
 
 
 
 
 
 
6bf9172
1aebf33
6bf9172
1aebf33
 
6bf9172
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
---
title: Redrob Candidate Intelligence
emoji: "πŸš€"
colorFrom: green
colorTo: blue
sdk: docker
app_port: 8501
pinned: false
license: mit
---

# Redrob Candidate Intelligence β€” Stage 1 Sandbox

This is the **Stage 1 sandbox** for the Redrob hackathon submission
(spec Section 10.5).

The Streamlit app accepts a candidate sample (≀5,000 records) uploaded
by the user, runs the **complete pipeline end-to-end** (BM25 + 5-axis
PageRank + 74 features + deterministic composite + reasoning), and
produces a downloadable ranked CSV.

For the **full 100k run**, the official reproduce command is:
```bash
python scripts/run_ranking.py --no_dense --no_lgbm
```

## Source

Complete submission with all 11 iterations (v1 β†’ v10 β†’ v11):
**https://github.com/AdityaProCoder/JobRanker**

## Files in this Space

- `app.py` β€” Streamlit sandbox UI (the entry point)
- `redrob/` β€” full pipeline package
- `scripts/run_ranking.py` β€” the official full-pool pipeline
- `requirements.txt` β€” Python dependencies
- `Dockerfile` β€” Docker deployment (this Space)