// while(alive) { keep_systems_up(); }

Akshit Malik

> ▊

Site Reliability Engineer @ Akamai — keeping a 20+ GiB/second Web Security Analytics platform fast, observable and boring (the good kind). 5+ years of Kubernetes, cloud, AIOps and automation across AWS & Azure.

On-call ready ⏱ uptime: … 📟 MTTR-obsessed 🤖 AIOps builder
deploy --view-projects curl -O resume.pdf book 1:1 session
▼ scroll

$ grafana-cli dashboard open impact.json

Impact Dashboard

Real numbers from production. No vanity metrics.

0%
SLA adherence, multi-account envs
0%
Faster deployments after client-migration re-architecture
0%
Less troubleshooting time via Python/Bash automation
0%
Manual config-map triage eliminated by my Topology Mapping Agent
0+ GiB/s
data ingested by the platform I keep reliable at Akamai
$0k+
AWS cost saved upgrading 100+ EKS clusters
0+
Helm charts authored & maintained
0+
microservices operated on EKS/AKS
0%
less DevOps toil via promotion-automation pipeline
24/7
production support across platforms

$ git log --career --oneline --graph

Career Pipeline

Every stage passed. Zero rollbacks.

▲
CURRENT · PROD

Site Reliability Engineer II @ Akamai

Sep 2025 → present · Remote

  • Own SLIs/SLOs & error budgets for a customer-facing Web Security Analytics platform ingesting 20+ GiB/second; SME for high-severity incidents (escalations, paging, postmortems).
  • Built an end-to-end SECESC agent — AI orchestration on GPU servers that automates incident triage and cuts on-call toil.
  • Engineered a Topology Mapping Agent (TMA) that maps live ingestion topology, resolves Kafka consumer↔app ownership & lag — killing 100% of manual config-map spelunking during triage.
  • Shipped an SRE agent for PagerDuty triage & remediation, improving alert signal-to-noise.
  • Grafana dashboards + alerting engineered to move MTTD & MTTR, not just look pretty.
KubernetesKafkaGPU/AIGrafanaPythonMCP
●
STAGE 2

Site Reliability Engineer @ Innovaccer

Oct 2023 → Aug 2025 · Noida

  • Ran 50+ containerized services on AWS EKS & Azure AKS with ArgoCD GitOps; Ingress NGINX for service-to-service traffic.
  • Authored 100+ Helm charts; automated ops with Python + Rundeck + CI/CD (Jenkins, GitLab, GitHub Actions).
  • 99%+ SLA across multi-account environments; promotion-automation pipeline cut DevOps involvement by 25%.
  • Upgraded 100+ EKS clusters to latest Kubernetes — saved $40k+; secrets hardened with HashiCorp Vault.
EKS/AKSArgoCDHelmVaultRundeck
■
STAGE 1

DevOps Engineer @ Unthinkable Solutions

Oct 2021 → Oct 2023 · Gurugram

  • Deployed 10+ end-to-end monolithic applications with automated build-test-release pipelines on ECS (Fargate) & EKS.
  • Zero-touch Jenkins deployments for monolith and distributed architectures.
  • Security hardening: SonarQube, ECR image scanning, ModSecurity rules on Nginx.
  • Defined KPIs, monthly metrics & escalation processes that kept SLAs green.
ECSJenkinsSonarQubeNginx
🏅 AWS Certified Solutions Architect '23
🏅 HashiCorp Certified '23
🎓 B.Tech CS & IT, AKTU — 85%

$ kubectl get projects -n production --show-labels

Flagship Projects

Written up like postmortems — because that's how SREs tell stories.

IMPACT: HIGH#PROJ-001

SECESC Agent — AI-orchestrated incident triage

runtime: GPU servers · stack: AI orchestration, LLM, MCP, Python

Problem
High-severity Web Security events demanded slow, manual, repetitive triage from on-call engineers.
Fix
Built an end-to-end agent that ingests events, correlates context and produces triage decisions automatically — deployed on GPU infrastructure.
Result
Dramatically reduced triage time and on-call toil; faster first-response on the events that matter.
resolved · running in prod
IMPACT: HIGH#PROJ-002

Topology Mapping Agent — know your ingestion flow, instantly

stack: Kafka, Kubernetes, Python · consumers ↔ apps ↔ lag

Problem
Finding which Kafka consumer belonged to which ingestion app meant manually digging through config maps — on every single triage.
Fix
End-to-end agent that auto-maps live ingestion topology, resolves consumer ownership and surfaces lag per component.
Result
100% of manual config-map inspection eliminated from initial triage. Minutes → seconds.
resolved · running in prod
IMPACT: MED-HIGH#PROJ-003

SRE Agent for PagerDuty — alert triage & remediation assist

stack: PagerDuty API, Python, AIOps

Problem
On-call fatigue: noisy pages, repetitive first-steps, context scattered across tools.
Fix
An agent that assists PD alert triage and remediation — gathers context, suggests next actions, executes safe runbook steps.
Result
Better signal-to-noise, less manual effort per page, calmer on-call rotations.
resolved · running in prod
IMPACT: MED-HIGH#PROJ-004

EKS Fleet Upgrade — 100+ clusters, zero drama

stack: EKS, Jenkins, Helm · scale: 100+ clusters

Problem
A fleet of 100+ EKS clusters drifting toward end-of-support — and AWS Extended Support bills.
Fix
Jenkins-orchestrated upgrade waves to the latest Kubernetes version with validation gates.
Result
$40k+ saved, fleet current, no customer-facing downtime.
resolved · fleet healthy

$ helm show values akshit/skills

Skill Matrix

☸ kubernetes:

KubernetesHelmArgoCDIngress NGINXCert ManagerOperatorsCrossplaneDocker

☁ cloud:

AWS (EKS · ECS · EC2 · VPC · RDS · S3 · Fargate)Azure (AKS · EntraID · Databricks)GCP

📈 observability:

PrometheusGrafanaDatadogOpenTelemetryELK/OpenSearchNew RelicCloudWatchFluentd

🤖 ai-infra:

GPU cluster opsML/AI inference deploysCI/CD for MLLLM/Claude · MCPAI orchestrationGPU capacity & cost planning

🔁 ci/cd & iac:

GitHub ActionsJenkinsGitLab CIArgoCDTerraformHelmRundeck

🌐 net & protocols:

Linux (strong)TCP/IPDNSLoad balancinggRPCWebSocketOAuth / OIDC

🐍 code:

PythonBashDSASystem DesignDistributed Systems

🔐 security & data:

HashiCorp VaultIAMWAFSonarQubePostgreSQLMongoDBRedisKafka-MSKSnowflake

$ ssh visitor@akshit.dev

Talk to my Terminal

A portfolio that says it's for an SRE should prove it. Type help to start.

visitor@akshit-portfolio: ~
visitor@akshit:~$ 

$ cat /etc/akshit/interests.conf

Beyond the Pager

🏏

Cricket

Strategy, teamwork under pressure — basically incident response with a bat.

✈️

Travel

New places, new cultures — multi-region redundancy for the mind.

📷

Photography & Food

Patience, precision, and appreciating the details — every photo and meal tells a story.

🏃

Athletics

Running, hiking, temple visits — keeping my own error budget healthy.

🌌

Curious Learner

From the universe to insects to history — curiosity never hits its rate limit.

🎙

Mentorship

I mentor engineers 1:1 on DevOps/SRE careers. Book a session →

$ echo "let's build something reliable" | mail akshit

Open a Connection

Open to SRE/DevOps roles, collaborations and mentorship. Handshake guaranteed — no timeout.

✉️SMTPcareer.akshit@gmail.com 💼LINKEDIN/in/careerakshit 📱WHATSAPP+91 97191 29353 🎯MENTORSHIPtopmate.io/akshit_malik
status: open to SRE III opportunities · response SLO: < 24h