Engineering Infrastructure
at Scale.

I'm Yasser. I build the cloud, ML and reliability layer behind products that suddenly need to handle real traffic. Head of Infrastructure at Barakah; previously ran the platform behind 400M+ daily requests for 70K merchants at Salla.

Talk about your stack ->See what I've shipped
platform - production
live
req/sec
412
peak 5,200 - 400M/day
p99 latency
184ms
SLO 250ms - within budget
spend trim
20.4%
karpenter - spot strategy
13:41:01argocdsync OK - prod-eks-1 - 14 apps healthy
13:41:09karpenterscaled spot pool +6 nodes - saved $3.20/hr
13:41:34istiop99 200ms -> 184ms (canary 25%)
14:10:57sagemakerqwen3-vl batch - 1,420 SKUs categorised
14:22:34guardimage scan clean - cosign verified
About Me

How I Think About Infrastructure

Every system I touch gets evaluated through four lenses: scale, security, cost, and business impact.

Yasser Alattas — Cloud Infrastructure Engineer
Open to senior roles

I'm Yasser Alattas — Head of Infrastructure at Barakah, based in Jeddah. I lead the cloud, data, and reliability platform behind Saudi Arabia's surplus-food marketplace, after running the compute platform for one of the region's largest e-commerce ecosystems at Salla.

Jeddah, Saudi Arabia
Currently @ Barakah
7 years in cloud infrastructure

The Path Here

I started at Puritec — containerizing apps, wiring up CI/CD, and managing infrastructure across AWS and Oracle Cloud for clients in consulting, staffing, and food-tech. Five years of that taught me how to build things that work across very different business contexts.

Then I joined Salla and the scale changed completely. I helped architect the EKS platform — multi-cluster Kubernetes with Istio, ArgoCD, and Karpenter — that serves hundreds of millions of requests daily. I also led Terraform adoption across five infrastructure domains: compute, storage, data, cybersecurity, and AI/ML.

Before leaving, I built Salla's MLOps platform — product categorization using vision and embedding models, inference pipelines on SageMaker, vector search with Pinecone. I learned GenAI alongside an ex-META mentor, which has been one of the best decisions I've made.

Right now, I'm leading infrastructure at Barakah, where I've been building since 2022. Kubernetes, Kafka, and infrastructure-as-code underneath; end-to-end observability from the app to the backend; GitOps with Argo CD; and a security and PCI DSS program on top — for a marketplace serving 2M+ users across 7 Saudi cities.

How I Work

Infrastructure should pay for itself

I think about cloud spend the way a CFO thinks about budgets. Every resource justifies its cost or gets cut.

Security isn't a phase

It's in the pipeline from day one. Scanned, signed, policy-checked — before anything hits production.

Build for the next 10x

Today's architecture should handle tomorrow's traffic without a redesign. I've been through enough scaling emergencies to plan ahead.

Ship things that matter

I don't build infrastructure for the sake of it. Every system I work on ties back to a business outcome — revenue, reliability, or speed.

Want to talk infrastructure?

I'm always up for a good conversation about scaling challenges, MLOps, or cloud architecture.

Reach OutSee My Timeline
Professional Journey

Where I've Built Things

From multi-cloud consulting to platform engineering at Middle East e-commerce scale — and now leading infrastructure at Barakah.

Head of Infrastructure

BarakahFoodTech Marketplace
Feb 2022 – PresentOngoing

Own the cloud, data, and reliability platform behind Barakah — Saudi Arabia's surplus-food marketplace connecting restaurants, bakeries, and groceries with customers.

Key Achievements

  • Built the cloud and data platform from the ground up — Kubernetes, service mesh, and Kafka event streaming, fully defined as infrastructure-as-code — behind a marketplace serving 2M+ users and 4K+ F&B partners across 7 Saudi cities
  • Run GitOps delivery with Argo CD across production and dev Kubernetes clusters, hardening workloads with disruption budgets, autoscaling floors, and redesigned health probes
  • Shipped end-to-end observability — distributed traces and metrics from the mobile app through the gateway to backend services, on a self-hosted telemetry stack — backing a 99% availability SLO
  • Lead security and compliance: turned an external penetration test into a prioritized remediation plan, mapped 86 PCI DSS v4.0.1 infrastructure controls to evidence, and centralized secrets management with isolated per-environment access
  • Head the infrastructure and Data & AI teams, and introduced FinOps (cloud cost visibility, CI and third-party API spend guardrails) plus a blameless postmortem practice

Technologies

KubernetesGitOps / Argo CDKafkaInfrastructure as CodeObservabilityData PlatformSecurity & ComplianceFinOps

Senior Cloud Engineer — MLOps Focus

SallaLeading E-commerce Platform
Dec 2023 – Aug 20262 yrs 9 mos

Owned the compute platform and MLOps infrastructure for Salla's e-commerce ecosystem.

Key Achievements

  • Architected the EKS compute platform with Istio, ArgoCD, and Karpenter — serving the platform's full API traffic across distributed microservices
  • Built product categorization pipeline using Qwen3 vision and embedding models with Pinecone vector database, enabling automated classification for 70K merchant catalogs
  • Led Terraform automation across 5 domains — computing, storage, data, cybersecurity, and AI/ML — governing the SDLC for all platform developers
  • Designed Karpenter spot-instance strategy that cut EC2 spend by 20% without reliability tradeoffs
  • Built SageMaker inference pipelines and multi-agent systems for production ML workloads

Technologies

Kubernetes (EKS)TerraformAWSIstio / ArgoCDKarpenterMLOps / SageMakerPrometheus / GrafanaPython

Senior DevOps Engineer

PuritecTechnology Solutions
Nov 2018 – Dec 20235 years

Led the DevOps function across AWS and Oracle Cloud, managing 5 production applications on Kubernetes for clients in consulting, staffing, and food industries.

Key Achievements

  • Deployed 5 production applications on Kubernetes with automated CI/CD pipelines via GitLab, standardizing deployment strategies across all projects
  • Managed multi-cloud infrastructure across AWS and Oracle Cloud, serving clients in management consulting, on-demand staffing, and food industry
  • Mentored junior engineers, wrote operational runbooks — on-call incidents dropped 30%
  • Expanded Amazon Partner Network (APN) relationship, unlocking co-sell opportunities and cloud credits worth $20K+

Technologies

AWSOracle CloudKubernetesDockerTerraformGitLab CI/CDPythonLinux

Interested?

If you're dealing with scaling challenges, cloud cost problems, or need someone to build your ML infrastructure — let's talk.

Get in Touch
In Production

What I've Shipped

Numbers from systems I've built and operate in production — not benchmarks, not demos.

5
Infrastructure Domains
Compute, storage, data, cybersecurity, and AI/ML — all managed through Terraform
20%
EC2 Cost Reduction
Karpenter spot-instance strategy at Salla — same reliability, less spend
30%
Fewer On-Call Incidents
Runbooks, SLO definitions, and mentorship program at Puritec
3
ML Pipelines in Production
Vision models, embedding models, vector search — product categorization at scale
$20K+
Cloud Credits Unlocked
Grew Amazon Partner Network relationship at Puritec into co-sell opportunities
2
Cloud Providers
Production infrastructure across AWS and Oracle Cloud

Where I Go Deep

Three areas where I have real production battle scars.

Platform Engineering

I build platforms that other engineers build on. Not just "keep the lights on" infrastructure — actual developer platforms with self-service deployments, observability built in, and autoscaling that doesn't need babysitting at 3 AM.

EKSIstioArgoCDKarpenterTerraform

MLOps & AI Infrastructure

Taking ML from notebook to production is a different game. I build the inference pipelines, model serving infrastructure, and vector search systems that make AI actually work at scale — not just in a demo.

SageMakerPineconeLangGraphQwen3LLMOps

DevSecFinOps

Security and cost aren't afterthoughts — they're architectural constraints I design around. Every PR gets scanned, every resource gets a cost tag, every deployment decision has a business justification.

TerraformZero TrustFinOpsIAMCloudflare

If your infrastructure is the bottleneck — I can help fix that.

Let's Talk

Certifications

AWS Solutions Architect — Associate

Cloud architecture design and AWS best practices

Get In Touch

Let's Talk

Have an infrastructure problem? Building an ML platform? Just want to nerd out about Kubernetes? I'm here.

Reach Out

I like solving hard problems — especially when they involve scale, cost, or making AI work in production. Drop me a message with what you're working on and I'll come prepared with ideas, not a sales deck.

Email

[email protected]

LinkedIn

linkedin.com/in/yasseralattas

GitHub

github.com/yalattas
Selectively available

Selective consulting slots open — response within 24 hours.

Send a Message

Tell me what you're building and what's not working. I'll get back to you within a few days.

Prefer a call?

Sometimes it's easier to just talk. Book a time and we'll figure out if I can help.

Schedule a CallSend Email