Shovel · AI-driven research & data acquisition · est. 2026

Custom research,
AI-driven and
human-verified.

ByShovel is an AI-driven research studio. When the answer is buried in public records, permits, filings, and agency data that nobody has organized, we go get it, structure it, and analyze it. AI does the reading at a scale a person can't match. A person checks the findings and the sources before they reach you.

01 · Proof

The method,
shipping live.

All products →

Mining Incidents

live

Searchable research tool for US mining safety records.

~500k records · 1983–present · refreshed weekly
What it does
What it does

Every reportable mining incident MSHA has on file — fatal and non-fatal — searchable by mine, operator, year, severity, or a keyword in the investigator's narrative. Built for safety inspectors training crews, researchers chasing patterns, journalists chasing stories, and families trying to find a record. Updated weekly from MSHA's open datasets.

Why we built it

I took an MSHA training course taught by Kim Redding and walked out unsettled. Every incident was on the record, but the public couldn't really find it — the agency's own search makes the data hard to see. This is a small counterweight: the same public record, opened up. Credit for the spark belongs to Kim.

How it works
  1. Pull the latest MSHA Accidents/Injuries/Illnesses snapshot from msha.gov each Monday.
  2. Normalize the schema and join against the Mine and Operator reference tables so every row carries operator name, mine name, and state.
  3. Run the investigator's-narrative field through an LLM classifier for incident category and severity.
  4. Index the result into Postgres with full-text search on the narrative field.
  5. Publish to miningincidents.org; preserve the original MSHA report ID end-to-end so every record traces back to the source row.

The Vault Report

live

Precious metals intelligence.

COMEX + LME ticks · warehouse flows · refreshed daily
What it does
What it does

COMEX and LME pricing and warehouse flows — synthesized daily into briefings, charts, and a fully automated video pipeline.

Why we built it

Gold coverage is either Bloomberg terminals (expensive, institutional) or YouTube stackers (vibes). There's a gap between those two for someone who just wants the feed: prices, flows, what the internet is saying, every morning. That's the Vault Report.

How it works
  1. Poll COMEX (CME Group) and LME tick feeds at 1-second cadence for gold, silver, platinum, and palladium.
  2. Join price action against COT positioning and on-chain warehouse flows into a single daily briefing each morning UTC.
  3. Render the briefing into a chart pack and feed the chart pack into the automated video pipeline.
  4. Publish to thevaultreport.com and push the briefing to subscribers.
Try a search like

Safety Incidents

live

Searchable research tool for US workplace-safety records.

~5.2M OSHA inspections · 105k severe-injury reports · refreshed daily
What it does
What it does

Every OSHA workplace-safety record on the federal file — severe-injury reports, inspections, citations, and accident investigations — searchable by employer, state, hazard, or industry. Built for safety managers checking a contractor, workers checking an employer, journalists chasing a pattern, and researchers who need the raw record. Refreshed continuously from the US Department of Labor's OSHA datasets.

Why we built it

MSHA covers mines; OSHA covers everyone else — every warehouse, plant, and job site in the country. Those severe-injury and inspection records are all public, but OSHA's own tools are built for compliance officers, not the public. Safety Incidents is the same federal record, opened up: type an employer, see their history.

How it works
  1. Pull OSHA's Severe Injury Reports, inspections, citations, and accident-investigation datasets from the US Department of Labor's DOL API on a rolling refresh schedule.
  2. Normalize each dataset into a common schema and resolve raw employer names into canonical establishment pages.
  3. Index everything into full-text search over employer, state, hazard, industry, and the incident narrative.
  4. Publish to safetyincidents.org; preserve each record's OSHA identifiers so every row traces back to the federal source.

MyBootLab

live

On-demand rebound-boot fitness classes, in Spanish.

live since July 2026 · new classes weekly · es-MX, en-US, pt-BR
What it does
What it does

A members' portal for rebound-boot classes — the spring-loaded boots used for low-impact bounce cardio — sold as recurring Lab Passes to a Spanish-speaking, mobile-first audience. Each coach gets her own catalog and her own storefront under one roof, so a member buys a pass for the coach she trains with rather than for a generic library. The public guides on boot safety, class formats, and how an online class actually works are free and need no account.

Why we built it

Rebound-boot classes are taught almost entirely in person, by independent instructors who have no way to sell to the students they already teach once class ends. The vocabulary is a second problem: the well-known brand names are trademarked, so the generic terms people actually search — botas de rebote, and Hopping in Brazil — belong to nobody and are answered mostly by unsourced marketing copy. MyBootLab is the portal those instructors can sell through, and the reference that answers the questions honestly.

How it works
  1. A coach records a class on her phone and uploads it straight to the video platform, never through our servers.
  2. The class is transcoded, its length measured from the file rather than guessed, and it appears in her catalog the moment it is playable.
  3. Members buy a Lab Pass for that coach's lab; every paywalled surface resolves through one indexed entitlement check.
  4. Watch time is credited from the player's own clock, capped per heartbeat, as the evidence a coach's future payout is computed from.
mybootlab.com
02 · Datasets

The ground truth
underneath every product.

Every product sits on top of a live dataset that Shovel collects, cleans, and keeps fresh. Below: what's growing in the pit.

MSHA Accidents & Fatalities

Refreshed weekly

Every injury and fatality reported to the Mine Safety and Health Administration — normalized, searchable, kept fresh.

Coverage · 1983 — present

Precious Metals Tick Feed

1-second polling

COMEX and LME feeds for gold, silver, platinum, and palladium — second-by-second.

Coverage · Live
03 · Studio

The
factory
owner.

Shovel runs three live US public-records databases: miningincidents.org (~500,000 MSHA mining-incident records, 1983 to present, refreshed weekly), safetyincidents.org (millions of OSHA inspections, severe-injury reports, and accident investigations, refreshed daily), and thevaultreport.com (daily COMEX and LME tick data plus warehouse flows). I design the playbooks and review what comes out; AI pipelines run them: fetch the source feed, normalize the schema, classify the narrative, index, publish.

The same machine runs on assignment. When an organization needs data that nobody has organized yet, scattered across agencies, permits, filings, and public records, the pipelines go and get it, normalize it into a clean schema, and analyze it, and I check the findings against the sources before they go out. Whether the question comes from a county, a law firm, or a newsroom, the work is the same. The three databases above are proof the method works: the same pipelines fetch from MSHA, OSHA, CME Group, LME, and Twitter every day, cleaned and put behind a public search box, used by safety inspectors, journalists, researchers, and families.

About the studio →

Have a dataset you need built? Start a research engagement →

04 · Build with us

Bring us the app
you wish existed.

Every product above is software we designed, built, and run: search tools, dashboards, APIs, pipelines that refresh every day. We build the same for other people. If you have an app idea, a tool your team needs, or a project you want a partner on, describe it here. We'll scope it and email you a quote, usually within a couple days.

Prefer email? hello@byshovel.com

05 · FAQ

Questions
and answers.

Can you find and build a dataset for us?

Yes. Custom research and data acquisition is what ByShovel does: if the answer you need is scattered across public records, permits, government sources, or filings and nobody has organized it, we go get it, structure it into a clean schema, and analyze it. AI does the reading at a scale a person can't match, and I check the findings against the sources before they reach you. The live databases here are proof of the method. Email hello@byshovel.com with what you're after.

Can you build an app for us?

Yes. The three live products here are apps we designed, built, and operate end to end: search interfaces, dashboards, APIs, and the data pipelines underneath. We build the same for clients, from an internal tool to a full public product, and we're open to partnering on the right idea. Describe your project in the form above or email hello@byshovel.com.

Where can I search MSHA mining incident reports?

miningincidents.org — every reportable MSHA mining incident from 1983 to present (~500,000 records), searchable by mine, operator, year, severity, and a keyword in the investigator's narrative. Refreshed weekly from MSHA's open datasets.

Where can I read a daily precious-metals report?

thevaultreport.com — COMEX and LME tick data for gold, silver, platinum, and palladium, plus COT positioning and warehouse flows. Refreshed daily into briefings, charts, and a video pipeline.

Can I license the underlying datasets?

Yes, by arrangement. The full corpus (MSHA + precious metals) is licensable for LLM training and other commercial use, priced per use case and scope. Email hello@byshovel.com with what you need, or see byshovel.com/licensing. Agents that need rows now, not a license, can buy per-call over x402 at each product's /api/x402.