# About hecigo

hecigo is an integration lab operated by **HECIGO TECHNOLOGY CO., LTD**, a
company registered in Ho Chi Minh City, Vietnam. We do exactly one thing: the
layer in between.

A 20-to-200-person company usually has enough software already, and each piece
does its own job well. What is missing is whatever makes them talk to each other,
and no vendor takes responsibility for that part. That gap is the whole of our
business.

## An integration lab, not a contractor

A contractor takes a specification and returns a build. A lab starts from the
position that nobody, including us, can quote a middleware build accurately
before looking at real data, because every pair of systems fails in its own way
and most of that lives in what the API docs never mention.

So the engagement is gated rather than quoted up front:

1. **Integration discovery**, 1-2 weeks, paid. A written technical assessment
   complete enough for anyone, including another vendor, to quote from.
2. **POC on real data**, one flow, 20-50 transactions, against acceptance
   criteria agreed before a line of code is written.
3. **Production build**, 4-8 weeks per system pair, with a parallel run alongside
   the old process before anything is switched off.
4. **Managed operations**, monthly, or a handover to your own team.

You pay a small amount, watch it run, and only then decide whether to continue.
Every step has an exit, and the deliverables from a step you stop at are yours to
keep.

## What we are accountable for

- We never ask you to replace software that already works.
- A small POC comes first, with acceptance criteria written up front. If it fails
  them, we stop.
- Handover is complete enough for another team to take over, and that standard is
  written into the contract rather than promised verbally.

The POC acceptance criteria are deliberately unglamorous and measurable: every
transaction reconciled by hand, source against target; five events replayed with
the target record count unchanged; latency reported at both P50 and P95, read
from the logs rather than estimated; and the target deliberately cut off to prove
the backlog drains fully on recovery.

## What we have shipped

Production integrations with **Base CRM, Lark Suite and Odoo**: some under
contract, some while running those systems from inside the business. Two
open-source n8n community nodes live on npm with outside users:
`n8n-nodes-zalo-platform` and `n8n-nodes-firecrawl-v2`. Integration
infrastructure we have operated ourselves for months, which is where most of what
we know about failure modes actually came from.

We are a small team. We say so on purpose: the alternative is a capability deck
that implies otherwise and a first invoice that discovers the truth.

## Research. Prototype. Deploy.

Not a slogan, three real steps. Discovery to understand the data, a POC to prove
the approach holds, then production. The working stack is Node.js, PostgreSQL,
Docker, n8n and Grafana, chosen because they are ordinary enough that another
team can pick the system up without us.

## Where to go next

- [Developer resources](https://hecigo.com/developers/): the open-source nodes and machine-readable endpoints
- [Blog](https://hecigo.com/blog/): mechanics and failure modes, including ours
- [Contact](https://hecigo.com/contact/): email, phone, address, enquiry form