ELSAADOUNI // SYSTEM PORTRAIT
000%
M.ELSAADOUNI — SYSTEM PORTRAIT v3.0 STATUS: OPERATIONAL ABU DHABI · 24.45°N 54.37°E EST. 2009

Mohamed Elsaadouni.

IT Leader Father Open Source Advocate Health Insurance Technology

I keep health-insurance systems reliable, tune Oracle databases, and lead teams that treat uptime as a promise — not a hope.

Mohamed Elsaadouni
Current Role
Manager — IT Applications
Org
DAMAN · National Health Insurance
Base Station
Abu Dhabi, UAE +04:00
Open to select missions
operator@elsaadouni — zsh
0Years in production
0Projects delivered
0Roles · one mission
0First login
01 Profile

Engineer by trade. Father by design.

A biography of the person behind the uptime — told in log lines, not labels.

I run production systems the way a good parent runs a household: with patience, discipline, and a little automation. For more than a decade I've worked in national health insurance technology, where a slow database can delay real people. That context shapes everything I build.

My toolkit includes Linux administration, Oracle databases, VMware and Hyper-V, and ITIL. I've led service desks, improved SLAs, built monitoring, and guided teams through deployments where "almost" was never good enough.

Off the clock I'm a father first and a technologist second. I tinker with home automation, try new Linux distributions, and contribute to the open source ecosystem that taught me so much. I believe good engineers stay curious, generous, and willing to read the docs.

— I still read the runbook before breakfast.
IDENTITY CARD● VALID
ME
NAMEMohamed Elsaadouni
ROLEManager — IT Applications
ORGDAMAN National Health Insurance
BASEAbu Dhabi, UAE
STATUS● OPERATIONAL
01 · The Father

Why uptime matters

Every 3 a.m. page, every weekend failover — there's a family at home. Priorities stay sharp.

02 · The Engineer

Systems that breathe

Infrastructure is a living organism: monitored, documented, automated, and respected.

03 · The Advocate

Share what works

I share scripts, notes, and the lessons that would have saved me time earlier.

02 The Journey

Five roles. One mission.

Seventeen years inside a single organisation's engine room — from the helpdesk floor to the command deck. Each role was a chapter; each chapter, a promotion in responsibility. Open a log to read the debrief.

The mission. Take a growing insurer's first and second-line support and turn it from a queue into a promise.

I started on the floor — headphones on, tickets flying. That's where I learned the first law of enterprise IT: users don't care about your architecture, they care about their problem being solved. Leading the helpdesk taught me to run a team like a well-ordered switchboard: clear escalation paths, honest status updates, and the discipline to close tickets properly instead of just closing them fast.

Those years built the reflexes I still operate on: when something breaks, you don't guess — you gather data, you communicate, and you fix the root cause, not the symptom. Every outage since has been a debt I paid down with the habits I formed at this desk.

1st / 2nd line supportEscalation designUser experienceService reliabilityTeam building
0Panic-buy tickets kept
24/7Support mindset installed
1stRule: fix the cause, not the symptom

The mission. Make performance visible. If it isn't measured, it isn't managed — and it certainly isn't loved.

This chapter taught me to speak fluent KPI and SLA. I built service level reports from raw application data, tracked uptime and response windows against contracted targets, and learned the quiet art of turning a spreadsheet into a decision. When a business owner asks "how are we doing?", the answer should be one glance away — not three meetings away.

The deeper lesson: numbers are a mirror, not a judge. A missed SLA tells you where the system is weak; a met SLA tells you where to invest next. I stopped treating reports as paperwork and started treating them as radar. Every dashboard I've designed since still runs on that principle.

KPI enforcementSLA reportingPerformance monitoringSQL analysisStakeholder comms
0Missed targets left unexamined
Glance to see the truth
100%Reports turned into radar

The mission. Own the daily operations of mission-critical applications — and sleep only when they do.

Five years at the sharp end: performance monitoring, configuration management, incident war rooms, and the kind of in-depth analysis that turns "something's slow" into "the redo log buffer needs attention". This is where I went deep on Oracle databases, Unix tooling, and the nervous system of enterprise applications — learning to read an AWR report the way a doctor reads an ECG.

I also learned the politics of prevention: convincing stakeholders to fund capacity before the crisis, not after, and building runbooks that turn a 2 a.m. page into a 15-minute resolution instead of a 3-hour scramble. The war room is won long before the alarm sounds — it's won in the documentation, the monitoring, and the rehearsals.

Oracle tuningConfiguration managementIncident war roomsCapacity planningRunbooksUnix / Linux
15 minTarget incident resolution
AWRFluency gained
0Fires left un-prevented

The mission. Stop being the best engineer in the room. Start building the room.

Promotion meant letting go of the keyboard — mostly — and taking responsibility for a team of specialists: their growth, their quality bar, and their sanity during deployments. I ran critical projects spanning application deployment and systems integration, where the technical work was the easy part and the orchestration of people, vendors, and deadlines was the craft.

The lesson that stuck: a leader's output is the team's throughput. I stopped asking "what did you do today?" and started asking "what's blocking you?". I protected the team from noise, gave them context instead of commands, and watched the same engineers who used to wait for permission start shipping on their own initiative.

Specialist team leadershipDeployment projectsSystem integrationVendor managementCoaching
TeamThroughput = my output
1stQuestion: what's blocking you?
Autonomy across the team

The mission. Steer the health insurance applications estate like a ship: every system charted, every crew trained, no icebergs unannounced.

Today I lead high-performance teams responsible for the vital health insurance applications that keep claims, policies, and provider networks running for one of the region's largest insurers. The brief is strategic: improve availability, modernise operations, and make sure every application under my watch meets the standard our members deserve.

My operating doctrine, distilled from seventeen years on the same mission: availability is a design decision, not an accident — layered monitoring, honest capacity planning, ruthless runbooks, and a culture where the post-incident review is a thank-you letter to whoever found the problem, not a blame trial. The systems are only half of it. The other half is a team that feels safe to tell you the truth.

Application strategyHigh availabilityPerformance teamsStrategic improvementITIL at scale
NinesAvailability as a design decision
24/7Members depend on us
TruthCulture over blame
03 Capabilities

Systems I run.

MODULE 01 / CORE SYSTEMS

The tools underneath the work

The foundation layer — where uptime is born. Measured in years of production, not certificates.

Virtualization — VMware · Hyper-V0%
Linux Administration0%
IT Service Management — ITIL0%
SQL & Oracle Database0%
MODULE 02 / COMMAND LAYER

Leading people, not just processes

A manager's real stack: communication, judgement, and the courage to let the team shine.

Team Leadership0%
Project Management0%
Strategic Planning0%
Security & Compliance0%
MODULE 03 / OPERATING VALUES

What I calibrate daily

Not soft skills — system invariants. These are the dials I check before coffee.

Uptime Disciplinethe nines are earned
Automation Biasscript everything twice
Security Firsttrust nothing, verify
Documentationif it isn't written, it didn't happen
MODULE 04 / INSTALLED STACK

What I use every day

The daily drivers — curated over seventeen years, updated continuously.

operator@elsaadouni: ~/stack
$

← Scroll to traverse the modules →

04 Field Reports

Case files from production.

A selection of representative engagements across health-insurance operations, automation, and open-source craft. Each report is written the way incidents should be: honest about the challenge, precise about the approach, measurable about the outcome.

Challenge

Insurance technology is a real-time promise: claims, eligibility and provider services run against a constant stream of transactions, and every minute of degradation lands directly in someone's healthcare journey. The estate spans databases, middleware and virtualized infrastructure — all of it expected to behave like a utility.

Approach

Layered monitoring with alerts tuned until they meant something. Runbooks for every known failure mode, war-room discipline for the unknown ones, and relentless Oracle/OS performance analysis to find the slow queries before the users did. Config management so every server looked like its twin.

Outcome

Availability measured in nines instead of anecdotes. Incidents resolved in minutes because the team had rehearsed the response. And a culture where "post-incident review" means improvement, not punishment.

High availabilityAWR / performance tuningRunbooksIncident managementCapacity planning
MTTR ↓Mean time to resolution
Nines ↑Availability discipline
Alerts →From noise to signal
Challenge

Performance data lived in application logs, monitoring tools and the heads of senior engineers. SLAs were negotiated in the dark and reported late. When a business owner asked "how are we doing?", the honest answer took three days to assemble.

Approach

Built a reporting pipeline that pulled metrics from source systems into structured SQL, automated the aggregation on a schedule, and rendered service level reports that arrived before the morning meeting. Every KPI traced back to a definition; every definition traced back to a contract.

Outcome

Decisions started from data instead of anecdotes. Target misses were caught in days instead of quarters — and the reports earned the team a seat at the table where service targets are set.

KPI / SLA reportingSQL automationData pipelinesStakeholder reporting
3 days →Report turnaround time
1 glanceTo see the full truth
Anecdotes ↓In decision-making
Challenge

Consumer smart-home ecosystems are walled gardens that phone home, break on a whim, and hold your house hostage to a vendor's roadmap. I wanted automation with real engineering standards: versioned config, monitored services, and zero data leaving the house.

Approach

Built a self-hosted stack on open-source components — containerised services, MQTT as the message bus, Home Assistant as the brain — with the same discipline I'd bring to a production estate: backups, documentation, and automation of the automations themselves. Every routine is a script; every script is in Git.

Outcome

A home that runs itself — lights, climate, sensors, and the occasional 2 a.m. tinkering session. More importantly: a permanent lab where new tools get battle-tested at low stakes before they're ever recommended for production.

Self-hostedPrivacy-firstVersion-controlled configAlways learning
0Data leaving the house
100%Open source stack
2 a.m.Tinkering hours logged
Challenge

Enterprise ops knowledge has a terrible habit of living in one person's head. When that person is on leave, the system blinks. The fix isn't a binder — it's a culture of writing things down and shipping the tools as reusable, shareable artifacts.

Approach

Published shell tooling, configuration templates and documentation as open-source artifacts: runbooks as living documents, scripts with real error handling, READMEs written for the next operator. Shared publicly where possible, internally where not — always versioned, always reviewed.

Outcome

Teams that can answer questions from the docs instead of from the phone tree. A public record of craft that has opened doors, sparked conversations, and made the case — inside and outside my organisation — that sharing is a strength, not a risk.

Shell toolingDotfilesRunbooks as codeKnowledge sharing
Phone tree ↓Answers from docs instead
0Questions left undocumented
Interest repaid
05 Operating Philosophy

Principles I don't negotiate.

The system is not the servers, the databases, or the dashboards. The system is the people who keep it alive — and the trust they extend to each other at 3 a.m.
P / 01

Uptime is a promise

Availability is designed, tested and rehearsed — never hoped for. If it can fail, there is a runbook; if there is a runbook, it has been rehearsed.

P / 02

Security is a habit

Not a project with an end date. Least privilege, honest patch management, and the reflex to ask "what's the blast radius?" before anything ships.

P / 03

Simplicity scales

The best system is the one the next engineer can understand at 3 a.m. Fancy is easy. Clear takes work, so I choose clear.

P / 04

Open beats closed

Standards, source, and knowledge want to be shared. Every walled garden I've met eventually became a maintenance burden.

P / 05

Measure everything

If we cannot measure it, we are probably guessing. Dashboards, SLAs and honest retrospectives — data is how we argue politely instead of loudly.

P / 06

People before tickets

Behind every ticket is a human with a deadline or a worry. Empathy is an operations strategy — it retains talent and calms incidents.

06 Engagements

Ways we can work together.

Advisory, operations, or build — each engagement starts the same way: an honest conversation about the current state, the pain, and what "done" actually looks like.

  • Database & query performance deep-dives (Oracle, SQL)
  • Application availability & health reviews
  • Capacity planning and upgrade roadmaps
  • Monitoring, alerting and runbook hardening
Fits: operations teams, CIOs Discuss a review
  • Server hardening & baseline configuration
  • Automation with shell tooling & config management
  • Backup, recovery and disaster-recovery drills
  • Performance tuning: kernel, filesystem, network
Fits: small IT teams, startups Talk servers
  • Fast, accessible, self-hosted web experiences
  • Internal tooling & dashboards that people actually use
  • Deployment pipelines & container workflows
  • Open-source-first technology selection
Fits: product teams, founders Sketch a system
  • Team structure, hiring and onboarding design
  • ITIL processes that survive contact with reality
  • Coaching senior engineers toward leadership
  • Culture work: blameless reviews, honest status
Fits: engineering managers, HR Build a team
  • Infrastructure security posture reviews
  • Access, privilege & credential hygiene audits
  • Patch and vulnerability-management gaps
  • Pragmatic remediation roadmaps (not fearware)
Fits: regulated industries, SMEs Audit my estate
  • Shell tooling, cron orchestration & job hygiene
  • Home & office automation on open-source platforms
  • Self-hosted dashboards and notification pipelines
  • Training teams to automate without fear
Fits: anyone tired of Monday-morning manual work Automate it

Engagement models

Advisory — a day, a week, or a standing retainer of honest technical counsel
Operations — hands-on leadership of critical applications and platforms
Build — architecture, automation and web experiences, delivered

Based in Abu Dhabi, available worldwide. English, Arabic, or that dialect of broken French IT people speak at conferences.

Every engagement starts with a 30-minute conversation. No decks or sales script — just the problem and a clear definition of done.
07 Pulse

Current status.

Here is the practical version: what I do now, where I work, and what I am learning next.

Abu Dhabi · Local Time
--:--:--GST +04
Timezone Asia/Dubai Uptime 17 yrs on the mission Status ● OPERATIONAL
ROLEManager — IT Applications · DAMAN National Health Insurance
BASEAbu Dhabi, UAE — remote-friendly worldwide
FOCUSHigh availability, performance culture, and applications that never apologise
LEARNINGKubernetes & observability tooling — plus whatever Home Assistant breaks this week
SIDE QUESTOpen-source tooling, dotfiles, and the eternal pursuit of the perfect runbook
AVAILABILITY● OPEN — select advisory & build engagements
Core systems Databases Network Home lab Coffee supply Documentation — always in progress
08 Open a channel

Say hello.

Whether it's a failing database, a team in need of leadership, or a smart-home idea that shouldn't involve the cloud — the channel is open. Expect a reply within 24 hours, usually faster.

contact@elsaadouni.com
Response window: < 24h
Timezone: GST (UTC+4) — I answer at odd hours anyway

Transmit a message

sent to my contact workflow · read by a human