█Swastik
Swastik
Agnihotri
|
DevOps | SRE | QA Automation Engineer
Founder of CrawlMindAI —
AI-Powered Automation SaaS
Building scalable, reliable & production-grade systems
Jump to any section — or scroll through everything
▲ IIT Guwahati — AWS+DevOps● Available
$ git push origin main
✓ Pipeline triggered — staged rollout initiated
✓ Health checks passing (3/3) — promoting to production
✓ Rollback trigger armed
$ kubectl get pods -n production
NAME READY STATUS RESTARTS
app-7d9f8b-xk2p1 1/1 Running 0
app-7d9f8b-mq9r3 1/1 Running 0
worker-6c4df9-lt7n2 1/1 Running 0
✓ All pods running — 0 failures
DevOps @ InfoZ IT Services
PG Cert: AWS+DevOps (IIT Guwahati)
Built: RAG chatbot w/ triple-redundancy AI
Open to: MLOps · SRE · Platform Eng
99.87%Uptime
40%Cost Saved
~0Deploy Fails
<5 minMTTD
DevOps EngineerInfoZ IT ServicesJul 2025 – Present
Technical LeadLDM CollegeOct 2023 – Jul 2025
SREHCL TechnologiesMar 2023 – Sep 2023
Frontend DevLDM CollegeJun 2022 – Mar 2023
Research InternIIT KanpurMay 2019 – Jul 2019
DevOps Engineer @ InfoZ IT Services
Rebuilt deployment pipelines with staged rollouts. Cut cloud costs by 40%. Reduced deploy failures to near-zero.
View Full Story →Cloud Infrastructure Automation PlatformDevOps
Provisioning 4 weeks → 15 minTerraformAWSGitHub Actions
Kubernetes CI/CD Deployment EngineDevOps
Deployments 10× faster — <30s rollbacksKubernetesArgoCDGitHub Actions
AI Agent Reliability System (In Progress)AI Infra (WIP)
ReAct pattern with multi-component failoverPythonLLMsObservability
AWSTerraformAnsibleDockerKubernetesArgoCDGitHub ActionsPrometheusGrafanaELK StackIAMRBACSecrets ManagementPythonBashFlaskPostgreSQLLlama 3.3RAG PipelinesSemantic Search
↓scroll for deep dives
// expertise
What I Do
Infrastructure that ships. Systems that stay up.
DevOps & Infrastructure
Pipelines. Containers. Zero downtime.
DockerKubernetesGitOpsCI/CD PipelinesSystem ArchitectureDevOps Automation
Near-zero
Deploy failure rate
- ✓ Rebuilt broken pipelines — near-zero failures
- ✓ Cloud costs cut 40% via right-sizing
- ✓ Environment drift eliminated via containerization
Data & Automation
From raw data to automated decisions.
PythonPostgreSQLDatabase OptimizationScrapySelenium
10,000+
Products tracked automatically
- ✓ 10k+ products — Scrapy+Selenium pipeline
- ✓ Exam latency 8s→1s — async batch writes
- ✓ Manual reports → automated pipeline
AI READY
AI Engineering
Local LLM. Zero hallucinations. Live context.
Llama 3.3 70BAI Agent DeploymentRAG PipelineAI/ML Infrastructure
60s
Market event to AI analysis
- ✓ Deployed 70B LLM locally — no API costs
- ✓ Context injection eliminates hallucinations
- ✓ RAG chatbot on this portfolio — try it →
Observability & Reliability
Alerts that mean something.
Alert noise reduced
200+→<20
MTTD improvement
hours→5min
Recurring SEV-2s fixed
3+/week→0
PrometheusGrafanaRoot Cause AnalysisPerformance TuningIncident Management
I fix root causes. Not restart scripts.
How I Think
→Fix the root cause
Not the symptom.
→Own the outcome
Architecture to 11pm incident.
→Fail loudly
Silent failures do not count.
Open to: DevOps · SRE · Platform · AI Infra
Loading projects...
// deep dive: career path
Where I've Worked
DevOps Engineer
@ InfoZ IT Services · Remote
July 2025 – Present
- Designed secure cloud infrastructure (IAM, RBAC, environment isolation) across dev/staging/prod
- Reduced MTTD from hours to <5 minutes by rebuilding monitoring and alerting (Prometheus, Grafana)
- Drove deploy failures to near-zero using staged rollouts and automated rollback triggers
- Automated 12+ runbooks, reducing manual ops time from hours to <3 minutes
- Cut cloud costs ~40% via right-sizing and scheduled scaling without impacting availability
DockerPythonBashGitHub ActionsPrometheus
View Full Story →// deep dive: engineering proof
CI/CD Pipeline
From git push to live in ~13 min — automated quality gates, GitOps deploy, and metric-triggered rollback. Zero manual steps.
// architecture: code → production
Click any stage to view the full implementation case study →
Pipeline Impact
Measured outcomes across the full delivery lifecycle
🚀Deploy Frequency
2×/month→20×/month
✓Deploy Failures
3 in 6mo→0
⚡MTTR
45–90 min→<2.5 min
⏱Pipeline Runtime
40+ min→~13 min
04. What's Next?
Get In Touch
Whether you have a project, a role you think I'd nail, or just want to say hi — I'll get back to you.
Direct Contact
swastikwork007@outlook.com
Available for opportunities