total
grand total
$0
spent · across 5 platforms · still counting
0
approx. tokens consumed
SRE DevOps AIOps Cloud Data & Ops
I keep production from catching fire, automate the boring parts, and occasionally let a large language model do my homework. Currently finishing a Master’s in Computer Science at the University of Cincinnati and looking for full-time roles where pagers go off and somebody wants them to stop.
I’m an SRE / DevOps / AIOps engineer with five-plus years of running cloud infrastructure that refuses to fall over. AWS, Azure, Kubernetes, Terraform, observability stacks, the occasional three a.m. PagerDuty page that turns out to be DNS. Always DNS.
Right now I’m wrapping up an MS in Computer Science at the University of Cincinnati while moonlighting as an AI Operations researcher at the P&G Digital Accelerator, where I build agentic LLM systems that are smarter than me on a good day.
I’m looking for full-time roles in Cloud, DevOps, AIOps, Data, Operations, and SRE. If you have a fleet of microservices, a flaky deploy pipeline, or a Grafana dashboard that nobody reads, we should talk.
P&G Digital Accelerator @ University of Cincinnati · Cincinnati, OH
UBA Solutions Pvt Ltd. · Kathmandu, Nepal
Cloudlaya LLC · San Francisco, CA
AWS (EC2, S3, Lambda, ECS, EKS, RDS, ASG, SNS, Secrets Manager) · Azure · GCP · Nginx · DNS · SSL/TLS
Jenkins · GitHub Actions · GitLab CI · ArgoCD · Terraform · Ansible · CloudFormation · Helm · Docker · Kubernetes
Datadog · CloudWatch · Splunk · ELK · Grafana · Prometheus · OpenTelemetry · PagerDuty · SLO/SLI
Chaos engineering · Incident management · Capacity planning · Error budgets · On-call · MOPs/SOPs/EOPs
LLMs · LangChain · RAG · Prompt engineering · TensorFlow · scikit-learn · SageMaker · Databricks · Vector DBs · MLOps
Python · Bash · JavaScript · Groovy · SQL · PostgreSQL · MongoDB · Redis
LLM-driven Android bug reproduction. Hooks into an emulator, screenshots with bounding boxes, and lets Gemini 2.5 Pro follow human bug reports.
End-to-end pipeline with Jenkins, SonarQube, Docker, Terraform and Helm. Blue-green and canary, container scanning, IaC, the whole production cosplay.
Health-check and alert validator with anomaly detection. Cuts false positives, tracks error budgets, routes incidents to Slack and PagerDuty.
Chaos toolkit for Kubernetes. Pod kills, network partitions, resource pressure. Ships with pass/fail SLO gates inside CI/CD.
A fully transparent, slightly embarrassing accounting of how much I’ve fed to large language models so they could autocomplete me into a better engineer. Token estimates are blended input/output at public list prices, give or take a few hundred million tokens. No, my parents do not know.
≈ 50M tokens
Heavy lifting. Architecture, refactors, and the answers I tell people I figured out myself.
≈ 167M tokens
The everyday workhorse. Pull requests, scripts, “why is this YAML lying to me”.
≈ 1.24B tokens
Bulk reasoning at suspiciously good per-token pricing. I asked, it answered, my CFO did not.
≈ 500M tokens
Multi-modal models reading PDFs, screenshots, and the occasional whiteboard photo from 2 a.m.
≈ 120M tokens
Powering CARBON, a published research project, and one year of “explain this paper to me at midnight”.
≈ 30M tokens
Since launch. ChatGPT Plus, Pro, the API, and one impulsive month of GPT-realtime.
* Token counts are blended estimates based on public list pricing for each model family. Actual numbers vary, but the receipts are real and increasingly difficult to explain.
University of Cincinnati · Cincinnati, OH
Cloud Computing · Advanced Algorithms · Machine Learning · Artificial Intelligence · Data Analysis
Tribhuvan University · Nepal
Operating Systems · Computer Networks · DBMS · Data Structures
Currently in Cincinnati, OH. Open to relocation, remote, or a Bat-Signal in the night sky.