Hi, I'mZiad Ahmedthat's me 👋Ziad Ahmed.I build systems thatscale.

Open to workSenior Software Engineer · 5+ yrs
Let's talk

Node.js · Go · Java/Spring Boot on PostgreSQL, MongoDB & Redis, running on AWS. And yes, I ship React, Next.js & React Native too.

Who's reading?
live trace
trace_id: 7f3a…10e
GET /v1/engineers/ziad-ahmed200 OK · 19ms
  • ├─ auth.verify role=visitor3ms
  • ├─ cache.get "experience" HIT1ms
  • ├─ postgres.query projects10ms
  • ├─ mongo.aggregate impact8ms
  • ├─ queue.publish "interview.requested"4ms
  • └─ respond { open_to_work: true }1ms
p99 < 50ms · error rate 0.00%

Companies I've built for

Production numbers, not side-project numbers.

Every figure below shipped to real users, on real infrastructure, with a real AWS bill attached.

Nugttah01
0×
faster Order List API

MongoDB index utilization restored — zero schema changes

Nugttah02
0%
less memory on exports

100,000+ record export pipeline, no timeouts

Nuxera03
0×
faster transcript rebuild

42 s → 0.3 s of event-loop time at the 120-chunk cap

Solid Ground04
~$0K/yr
cut from AWS spend

RDS upgrade, idle EC2s, unused Elastic IPs

Solid Ground05
0
REST APIs, solo-built

~85K lines of TypeScript · 629 commits · 9 months

Proven Solution06
804 → 1
project loads per save

~1,600 redundant queries and 10K log lines gone; the platform stopped stalling

I collaborate with teams to build systems that stay boring in production.

Boring is the goal: predictable latency, readable code, and an AWS bill nobody panics about.

My experience
  1. Senior Team Lead Software Engineer at Proven Solution
    AWS · DevOps · Performance
  2. Senior Engineer at Nuxera
    Healthcare AI
  3. Senior Backend Engineer at Nugttah
    Jul 2025 – Jul 2026 · Full-time
  4. Senior Backend / Platform Engineer at Solid Ground Solutions LLC
    Jan 2025 – Apr 2026 · Contract · Remote (US)
  5. Teaching Assistant at AASTMT
    Feb 2026 – Jun 2026 · Part-time
  6. Backend Engineer (Mid-level) at Botit
    Sep 2023 – Sep 2025 · Full-time
  7. Freelance Software Engineer at Sonbola (السنبلة)
    Freelance · Customs clearance · Alexandria
Based in
Egypt
local time --:-- · remote-ready
Education
B.Sc. Computer Science
Arab Academy for Science, Technology & Maritime Transport
3.72
GPA / 4.0
#3
in cohort
Excellent with Honor
On my bookshelf
Designing Data-Intensive Applications
Martin Kleppmann
The big ideas behind reliable, scalable, and maintainable systems
How I work
01 Prove the test catches the bug

“Verified it fails when reverted” shows up in my PRs again and again. A test that passes either way proves nothing.

Don't take my word for it. Run it.

Interactive toy versions of systems I've built in production. Break them, crash them, switch the algorithm and watch what happens.

t=0s
Acme Corp
bulk campaign
+6
wait0s
Bloom Dental
small office
queue empty
wait0s
Cedar Realty
small team
queue empty
wait0s
8 call workers

What you're seeing

One customer launches a huge campaign. With a single FIFO queue, the small customers wait behind it: they wait 0s and climbing.

Leader election · Redis + Lua

👑
node-1
leader
◦
node-2
follower
◦
node-3
follower
lock TTL
3000ms
  • node-1 › SET dispatcher:leader node-1 NX PX 3000 → OK

Four systems, the best of what I've shipped.

The rest of the wins are in Achievements below.

Secure, measured, honest with data

Backend core of a medical transcription and clinical-documentation platform used by doctors. I made it faster, closed its security holes, and stopped it from showing fake numbers.

  • Audio pipeline: async ffmpeg and one probe per chunk, for 3.7× throughput and 74–118 ms less event-loop blocking per chunk
  • Transcript rebuild at the 120-chunk cap: 42 s → 0.3 s of event-loop time, with Arabic-aware numeral matching
  • Security: closed four privilege-escalation paths and locked down unauthenticated voice-biometric endpoints
  • Honest analytics: every metric is live | partial | proposed, so the UI says “not instrumented” instead of a fake 0

Integrations: PostgreSQL · OpenAI · ffmpeg · Opus · ONNX · React Query

browser exttranscribe APIauth · roleswake wordffmpeg chunkschunk joinanalyticsPostgresCDI · coder
300
Commits
7
Repos
3.7×
Throughput
42s → 0.3s
Rebuild
Node.jsTypeScriptPostgreSQLReactffmpegONNX

Every win, in one place.

51 achievements from 6 companies, plus open source, teaching and awards. Filter by what you care about.

The 3am stories. Real incidents, handled.

Anyone can build the happy path. Four times production caught fire, and what I did, written as blameless postmortems.

All systems operational
13 incidents resolved · 0 open
2023career uptimetoday
    1. 500s everywhere→
    2. CloudTrail: no human change→
    3. Snapshot→
    4. Resized & restored
    What happened
    Production started returning 500s on every API.
    What I did
    Traced it to a memory-starved RDS MySQL instance and used CloudTrail to rule out any human change. Took a snapshot, then resized prod from db.t4g.micro to db.t4g.small.
    Outcome
    Service restored, with CloudTrail evidence that nobody changed anything.
9 more reliability & security wins in Achievements ↓

Backend at heart. Full-stack when it matters.

Node.js, Go and Java/Spring Boot on the server. React, Next.js and React Native when the product needs a face.

Languages
GoTypeScriptJavaScriptJavaPythonSQL
Backend
Node.jsNestJSExpressSpring BootSpring CloudRESTGraphQLgRPCWebSocketsWebhooksffmpegOpenAI API
Data
PostgreSQLMongoDBRedisMySQLSQLiteFlywayIndexingQuery optimizationSchema design
Cloud & DevOps
AWS EC2RDSS3SQSSNSSESALBCloudFrontCloudTrailAmplifyDockerKubernetesTerraformGitHub ActionsNGINXCloudflare
Messaging
BullMQSQS / SNSRabbitMQKafkaRedis jobs
Frontend & Mobile
ReactNext.jsReact NativeReact QueryBrowser extensionsTypeScript
Architecture
MicroservicesEvent-drivenClean ArchitectureDDDSOLIDThreat modeling
Testing
JestSupertestMochaCypress

for students & juniors

Learning backend? The path I'd take again.

I teach Software Component Design and Software Architecture as a TA at AASTMT. This is the short version of what I tell my students.

▶ KO-DB on YouTubeSource on GitHub ↗
  1. 01Go deep on one languageGo, Java or TypeScript. Learn its runtime, concurrency model and standard library before chasing frameworks.
  2. 02Learn SQL like it pays rentJoins, indexes, transactions, isolation levels. Read EXPLAIN output until it reads like English.
  3. 03Build a real HTTP APIAuth, validation, pagination, error handling, tests. Document it with OpenAPI, then deploy it.
  4. 04Add caching & queuesRedis, BullMQ or SQS. Learn retries, idempotency and what “exactly once” really costs.
  5. 05Ship to the cloudDocker, a VM or container service, a managed DB, logs and metrics. Then read your first AWS bill.
  6. 06Build a database from scratchA tiny KV store with a WAL and a TCP server teaches more internals than any course. (That's KO-DB.)

FAQ

The questions recruiters and hiring managers usually ask in the first call, answered up front.

Senior software engineering roles, backend-heavy or full-stack, where I own services end to end: API design, data modeling, queues, cloud cost and reliability.

🤝

Tell me about your next system.

Hiring for a senior engineering role, need a platform rescued, or just want to talk databases? My inbox is open.

✉ Email meWhatsAppLinkedIn
[email protected]