Safety management built by people who actually run rigs.
Pre-job hazard reviews, lessons-learned recall, and incident pattern detection — for drilling and construction crews who are tired of safety systems that exist for the auditor and not the worker.
Not a SaaS pitch. I run a drilling company. The safety stack we use today is duct-taped together from PDFs, Teams chats, and goodwill. This is what I'm building to fix it — and what your crew can pilot before everyone else.
Three things, done well
Not a dashboard. A working co-pilot.
Pre-Job Hazard Review
Crew sees the same hazards in 30 seconds the JSA tries to surface in 15 minutes. Pulls the last 200 incidents on this hole-section / structure-type, ranks the top 5 by today's conditions, and serves them on the safety meeting screen.
- ·Site, weather, crew, equipment all factor in
- ·Spoken-language summary at the toolbox talk
- ·Sign-off captured to the audit trail automatically
Lessons-Learned Recall
Three years of incident reports, near-miss notes, and ad-hoc lessons buried in PDFs across SharePoint. Ask the question in plain English: "Have we ever lost a tool joint on this rig?" Answer in five seconds, with the original page numbers cited.
- ·Indexes everything: PDFs, emails, photos, audio, Teams
- ·Cites the source so the safety officer can verify
- ·Knows when something is anecdote vs documented event
Incident Pattern Detection
The third dropped-object call on this section in 90 days isn't a coincidence — it's a pattern. The agent watches the stream of daily reports and flags clusters before the supervisor reading 40 dailies a day would have caught them.
- ·Detects clustering by location, equipment, shift, crew
- ·Routes the alert to the right person, not the inbox void
- ·No "someone forgot to file the form" failure mode
Why this exists
The current safety stack is built for the auditor.
On most jobs the JSA gets filled in the morning, signed by everyone, and never referenced again. The lesson from last year's near-miss lives in a SharePoint folder no one searches. The pattern across three crews and four months never surfaces because nobody is reading 40 daily reports a day looking for it.
That's not a process problem. It's a tooling problem. We need a co-pilot that does the reading, the recall, and the pattern-matching — and surfaces the answer at the toolbox talk, not in a quarterly report.
That co-pilot is what Egan Forge has been quietly prototyping. The same autonomous-AI building blocks that run our public crypto-signal business and our 17 free SaaS tools (eganforge.com), now applied to the work we actually know — drilling, construction, field operations.
Early access cohort
We're piloting with a small number of operators in 2026.
If you run a drilling, construction, or field-services crew and your safety system frustrates you — get on the early-access list. We'll talk before we sell.
Request a pilot conversationOr DM Josh on LinkedIn. First responses go to operators who can describe what their current safety system gets wrong today.
Part of Egan Forge · See live company metrics · All products
Trade Smarter with AI Signals
Multi-model AI consensus (Claude + GPT-4o + Grok) delivers high-conviction crypto signals to your Telegram. Every trade published. Starting at $9/month.