00Plate & serial
Software
with tonnage.
Founded 2019. Eleven engineers. No sales team, no bench, no junior program. This page is the nameplate: who we are, how we decide, and what we're rated to carry.
01The story
Built by people
tired of watching
projects fail.
Heavy Industries started in 2019 with three engineers and one grievance: the important systems kept failing for organizational reasons, not technical ones. Twenty-person vendor teams where four people worked. Estimates built to win deals, not to hold. Seniors in the pitch meeting, juniors in the repo.
So we inverted it. Keep the team small enough to skip status meetings. Staff only people who've carried production systems for a decade. Put every line of code in the client's repo from the first day, demo running software every Friday, and price the work so we can afford to tell the truth about dates.
Seven years on, the model holds: 31 engagements shipped, four became long-run partnerships, and two we declined mid-sale because the date was fantasy. We count the last two as wins.
02Operating principles
-
Senior only.
Every engineer here has at least ten years carrying production systems. Not because juniors are bad — because your deadline is not the place to train them.
-
Small teams win.
Three to five people who trust each other outrun a twenty-person program every time. Communication overhead grows with headcount; we keep headcount small.
-
The date is the spec.
Scope bends, dates don't. We'd rather ship a smaller system on the promised day than a complete one four months late — and we scope so that "smaller" is still useful.
-
Boring is a feature.
Proven tools, dull schemas, obvious code. We spend novelty where it pays — the matching engine, the scheduler — and nowhere else.
-
Your repo, your system.
We build in your infrastructure and plan our own exit. A dependency on us that outlives the engagement is a defect, not a revenue stream.
-
Bad news first.
Bad news ages badly. If the estimate is wrong, the architecture won't hold, or your plan has a hole in it, you hear it this week — with the math attached.
03Capabilities matrix
Rated loads.
| Languages | Rust · Go · TypeScript · Java/Kotlin · Python · SQL |
|---|---|
| Data & messaging | Kafka · Postgres · ClickHouse · Redis · S3/object storage · Flink |
| Infrastructure | Kubernetes · Terraform · AWS / GCP · bare metal when latency demands it |
| Interfaces | gRPC · REST · FIX · binary wire protocols · legacy mainframe integration |
| Practices | Shadow traffic · parallel runs · load harnesses · trunk-based CI · on-call we design, then hand over |
| Domains | Capital markets · freight & logistics · insurance · energy · industrial telemetry |