Hey, I'm Hitesh. I build cloud infrastructure products, and the loops my agents run on.
I work at E2E Cloud, in Delhi. A few things I like below, and what I actually work on further down.
Manchester City
Haven't missed a match in years, bad runs included. Still get properly nervous before the big ones.
The AI hardware stack
I follow the whole AI supply chain, from equipment makers and foundries through model labs to the application layer, trying to work out where the real value actually sits, and where I should put my own money.
Pickleball
Picked it up last year on a whim. Still not good at it, but I keep showing up most weekends anyway.
What I've been building
Software Engineer at E2E Cloud, May 2025 to present, Delhi, India.
Load Balancer as a Service (LBaaS)
HAProxy, Ansible, Django, Prometheus, Ceph → LINSTOR
I own this product now: backend, customer escalations, the roadmap, all of it. Shipped end-to-end SSL encryption, health checks, and routing changes that cut latency and pushed revenue up 25%. I also led a cross-datacenter migration of 100+ load balancer VMs onto an entirely different storage backend (Ceph to LINSTOR), upgrading the entire load balancing software stack (HAProxy, Prometheus, Zabbix) in the same cutover, and held actual downtime under 3 seconds.
AI-native cloud revenue, product growth & inventory analytics
Python, FastAPI, React, FastMCP
Sales, Ops, Finance, and leadership (50+ people) use this daily to track GPU/CPU capacity against revenue. It's a dashboard plus an MCP server with 46 tools, so people, and AI agents, can just ask it questions instead of digging through spreadsheets. It slices $10M+ of monthly revenue across 15+ dimensions, saves Finance, Sales, and Ops 100+ hours a month previously spent on manual reconciliation, and gives leadership real-time decisions off self-serve information instead of waiting on a report.
e2ectl (open-source CLI)
GitHubTypeScript, Node.js, Vitest
Pitched and built E2E Cloud's official CLI from scratch, covering nodes, VPCs, volumes, and security groups. Paired it with an agent skill so Claude and Cursor can provision infrastructure from natural language.
Cloud platform console backend
Django, Celery, RabbitMQ, Go
I work on the backend behind our cloud platform console: async provisioning workflows for VMs, storage, and networking products, plus contributions to our Terraform provider in Go for new cloud services. I also own our payment system, including credit management, and handle production tickets on VPC routing and security groups.
Stack
Languages: Python, TypeScript, Go, C++, SQL
Applied AI: Agent building, MCP servers, Claude skills, LLM tool use, RAG, inference engineering
Backend: FastAPI, Django, Celery, FastMCP
Infra & networking: HAProxy, Terraform, Ansible, Docker, Kubernetes, Linux
Data & ops: MySQL, PostgreSQL, Redis, Prometheus, GitHub Actions
LeetCode Knight: top 6% globally, 350+ solved, rank #1,278 in Contest 404. Profile
B.Tech, Computer Science: LNMIIT, Jaipur (2021 to 2025).
1st rank, Agra: across four SOF Olympiads (Computer Science, Math, Science, English).
Get in touch
Easiest by email. LinkedIn and GitHub work too.
Message sent
Thanks, I'll get back to you soon.