Selected builds, systems, and tradeoffs. Raintree product pages live at raintree.technology.
Selected projects
17 projects
Agent Safety & Evals(4)
Apps(4)
Context & Developer Tools(3)
Data & Intelligence Systems(1)
Product SaaS(2)
Systems & Low-Level(1)
Music & Creative Systems(2)
Raintree Minerals
Intelligence / personal case study
Source-cited market research
What I built
Federal capture intelligence that connects SAM opportunities to critical-minerals market signals.
daily triage
SAM.gov
intel catalog
58 minerals
market claims
Source-cited
capture ops
Desk
Stack
Next.js 16
Cloudflare Workers
Bun agents
Neon Postgres
Cloudflare D1
Drizzle
NVIDIA NIM
Runtime
Cloudflare Worker crons dispatch daily opportunities, weekly entity-drift, and weekly competitor-activity feeds into the same bearer-authenticated triage path.
Bun agent CLIs run mineral analytics, D1 sync, MOS scoring, SBIR pilots, SAM/EDGAR research, digests, backfills, and public-scope checks.
The `/desk` operator surface exposes bids, capture-rate metrics, backend loop traces, subs, mandates, capture memos, and intel dashboards.
Data model
SAM.gov opportunities, SAM Entity rows, USAspending awards, competitor observations, target agencies, and sub-network records feed the govcon store.
A 58-mineral catalog, MOS scores, prices, observations, catalysts, market-tightness signals, and public capability content feed D1 intel views.
Solicitations, organizations, awards, minerals, mandates, agencies, capture memos, backend-loop checks, and triage decisions connect through shared entity IDs.
Techniques
Triage rule v2 combines mineral criticality, MOS, market tightness, policy pressure, whitespace, timing, set-aside fit, and capture fit.