Muhammad Faheem Khan

Device Intelligence & Diagnostics Platform — 1B+ IMEI checks for carriers, OEMs & refurbishers (US, EU, China)

• AI/ML Platform (GPU Inference): Deployed computer vision inference for the AI-Grading product on EKS, provisioning Karpenter GPU spot nodes (g4dn.xlarge) to resolve cold-start pipeline failures; enabled sub-second cosmetic defect detection (MobileNetV2), boosting device grading throughput and margin accuracy for global enterprise clients.

• Agentic AI Pipeline: Architected a Jira-to-GitLab autonomous code-generation pipeline using API Gateway, Lambda, and AWS Bedrock Action Groups; orchestrated ticket context ingestion, implementation code generation, and automated merge requests to pioneer agentic development practices.
• EKS Migration, Observability & Cross-Region: Directed the migration of 5+ microservices (CloudPortal, Dash, Shopfloor, Loupe) from EC2/ECS to Terraform-managed EKS with ArgoCD GitOps; instrumented Datadog APM, RUM, and SLO alerting, slashing cross-continental US-EU latency via optimized CloudFront and Route 53 geoproximity routing.

Luxury Fashion & Retail — Coach, Kate Spade, Stuart Weitzman | $7.75B Revenue | 1, 400+ Global Stores

• EKS Platform & Build Optimization: Managed multi-account EKS environments using Python CDK and Karpenter; engineered a daemonless Buildkit image pipeline that cut container build and deploy times from 5 hours to 10 minutes, accelerating feature releases across global e-commerce channels.
• AI/ML & GitOps: Orchestrated ArgoCD GitOps for Airflow, MLflow, and Evidently AI; leveraged ApplicationSet controllers to automate deployments powering demand forecasting, inventory optimization, and customer analytics across omnichannel retail operations.
• Jenkins as Code (JCasC): Developed modular Jenkins Shared Libraries and specialized agents (CloudFormation, Lambda, ECS) for reproducible CI/CD, reducing pipeline configuration drift by 80% and supporting multi-brand deployment cadences.
• Modernization & Observability: Migrated 10+ business-critical legacy applications from EC2 to containerized EKS, improving deployment frequency by 3x; implemented Datadog APM and distributed tracing to maintain production visibility across brand platforms during cutover.

Senior SRE & DevOps Lead - Sleep Number Corp. (Smart Sleep & IoT | 300K+ SleepIQ beds | $1.3B Revenue)

• Infrastructure Redesign: Architected the AWS backbone for the SleepIQ IoT platform — utilizing device gateways, Kafka, ECS, Cassandra, and Aurora to ingest real-time biometric data from 300, 000+ connected smart beds; optimized infrastructure footprint to slash cloud expenditure by 22% via Cloud Custodian and rightsizing.
• Security, Observability & IaC: Configured Sumo Logic SIEM with AWS integrations (RDS, Lambda, EC2), building unified dashboards that cut root-cause analysis time by 25%; rewrote the global IaC base with reusable network, compute, and security modules, reducing environment build time by 30%.
• SRE Leadership: Unified SRE, Platform, and DevOps squads via structured monthly syncs, improving incident response time by 35% and enforcing SLO-driven reliability practices critical to maintaining 24/7 IoT cloud uptime commitments.

Distributed Systems Engineer - USIS Procon (Enterprise Workforce Management | Multi-tenant SaaS Platform)

• Workflow Orchestration: Replaced fragile batch jobs with Temporal.io (Golang) durable workflows, boosting pipeline fault tolerance by 60% and cutting operational incidents by 40% across enterprise workforce scheduling tracks.
• Hybrid Cloud & Compliance: Managed on-premises Kubernetes clusters with BGP peering and Helm; enforced CIS Benchmarks (Kube-bench) to ensure stringent multi-tenant security compliance.

Cloud & Infrastructure Engineer - MonoLets Inc. (IoT Asset Tracking | Bluetooth mesh, 100M+ assets)
• Cloud-Native Architecture: Built a high-concurrency API on GCP Cloud Run using Golang (Goroutines/Channels), handling 2M+ daily requests at sub-100ms p99; deployed Dataflow, Pub/Sub, and ScyllaDB pipelines to cut IoT data-sync latency below 3 seconds.
• Event-Driven Scaling: Provisioned GKE Autopilot with KEDA driven by custom message-bus metrics, supporting real-time location ingestion from hundreds of millions of Bluetooth mesh asset tags across logistics pipelines.

• Serverless & Automation: Developed AWS Lambda monitoring crawlers using Python/Scrapy, improving latency metric collection by 15%.
• IaC Deployment: Delivered high-availability three-tier architectures using CloudFormation, accelerating deployment cycles by 30%.