Who does it:
- Startup developers (1–20 person teams) — the primary candidate
- IT services exporters (infosys/TCS/Wipro model) with internal tooling teams
- Product companies in Bangalore, Hyderabad, Pune, and increasingly Jaipur and Ahmedabad running Rails/Django/Node stacks on AWS/GCP
- In-house dev teams at mid-market Indian enterprises (₹50–500 Cr revenue) who have no dedicated DevOps or Platform Engineering headcount
- Personal laptops + Homebrew/Chocolatey; no standardized dev environments
- GitHub/GitLab (free tier) for repos; GitHub Actions (free tier) for CI; no paid tooling
- WhatsApp groups for deployment alerts, code review pings, and incident coordination
- Google Sheets or legacy Excel on a shared drive for tracking API specs, feature flags, technical debt
- Postman (free) for API testing; no contract testing, no API gateway
- No dedicated SRE function; the "on-call" is whoever's laptop is open
- Jira for large orgs; Notion or Trello for small ones; Excel for nothing-that-matters
- Internal wikis that rot — Confluence nobody updates, Google Docs nobody finds
- Every prod incident wastes 2–4 hours because no structured runbook, no alert routing, no postmortem capture
- Developers manually run
kubectl logs,kubectl describe podon their terminal — no unified observability dashboard - New developer onboarding takes 3–7 days because environment setup is tribal knowledge on someone's laptop
- API contracts between frontend and backend teams break silently; QA finds it in testing; sprint burns
- Security scanning is "we'll do it before launch" — it never happens before launch
- CI pipelines built by one person break when they leave; nobody knows what the pipeline does