Loading page...Загружаем страницу...正在加载页面...
AtlasRepo

Tender Radar: finding government tenders for small contractors, entirely on their own laptop

Description and use

Tender Radar enables small contractors to locally assess government tenders, verify eligibility, flag risks, estimate win probability, and draft compliant bids using Gemma 4 and TabPFN.

Problem

Small electrical contractors need to quickly assess government tender notices for eligibility, compliance, and win probability, but lack tools to do this locally and securely.

Outcome

On an M3 laptop with 8 GB RAM, Tender Radar processed 8 tenders in 22 minutes, extracting 51 clauses, producing 24 met, 5 partial, 19 missing verdicts, and drafting a bid pack with 8 unsupported sentences removed. Win‑probability AUC improved from 0.731 (sklearn baseline) to 0.752 (TabPFN).

Workflow

  • Clone the repository
  • Generate synthetic demo corpus
  • Load corpus into local database
  • Scan tender notices
  • Open the dashboard to review board and details
  • Draft bid pack

Evidence

Tender Radar answers all three on the contractor's own machine: Extracts the eligibility clauses…

Source

Measured on an M3 laptop with 8 GB RAM, everything local: 51 clauses extracted across 8 tenders… 22 minutes for 8 tenders.

Source

Win model ablation on the same time‑ordered split (147 submitted bids): sklearn baseline AUC 0.731 / Brier 0.260 vs TabPFN v3.5 + thinking: AUC 0.752 / Brier 0.209.

Source

  • Record date: 2026-10-17T15:56:13.293Z