Hisham Hanawi: a generalist automation engineer with a reliability habit
I'm Hisham Hanawi, and Hanawi Systems is my operations automation practice. I build the operations systems a business runs on: workflow engines, data layers, internal tools and the integrations between them.
The range is deliberate. I have built hiring pipelines and retrieval systems for recruitment, voice agent applications, lead generation engines, PPC-driven acquisition systems and internal operations consoles. Automation problems look different across industries and turn out to be the same underneath: data copied by hand, follow-ups that depend on memory, two systems that disagree, spend with no ceiling. The method transfers, and the work is the evidence.
What never changes from project to project is how I build. Idempotent and safe to re-run, so a duplicate trigger never double-writes. Audit-logged, so every automated action is attributed and reversible. Human-gated, so the system asks a person whenever a wrong automated decision would be expensive. Documented, so someone other than me can operate it.
Track record
| 100% | Job Success ScoreUpwork, verified 2026-09 |
|---|---|
| 583 | Hours, 6 engagementsUpwork, verified 2026-09 |
| $410K+ | Client revenue, prior systemsPrior role, as stated on the Upwork profile |
| 34,000+ | Verified B2B contactsCareer total, rounded down |
| TR+ | Top Rated PlusUpwork, verified 2026-09 |
Across 6 engagements on Upwork I have delivered 583 hours, including a single long-term role of 524 hours. Before that, systems I built produced 12,000+ leads and over $410,000 in client revenue in under three months, with acquisition costs lowered by over 37 percent.
Clients on Upwork have endorsed the same traits across jobs: clear communicator, committed to quality, detail oriented, solution oriented.
The profile carries 100% Job Success and Top Rated Plus standing, with five stars on every job that was rated.
Platform figures verified against the Upwork profile, 2026-09. Prior-role figures as stated on the profile.
Five working principles
-
Map before build
The expensive mistakes happen when someone automates the org chart's version of a process instead of the real one. I map first, every time.
-
Idempotent by default
Everything is safe to re-run. Duplicate triggers, retries and mid-run failures are normal operating conditions, not emergencies.
-
Audit-logged
If a system writes data, the write is attributed and reversible. "What actually ran" should be a query, not an argument.
-
Human-gated where it counts
Automation earns trust by knowing its limits. Anything ambiguous or expensive stops and waits for a person.
-
Handover is part of the build
Documentation someone else can act on. You should never need me to understand your own operation.
The method is the product.
If your operation is held together by manual work and good intentions, I will show you what a system in its place looks like.