Technology owner and architect who runs production infrastructure, security and products across the Spoorthy group — setting direction, making the risk calls and holding every release to a high bar.
Owns and operates a multi-server AWS production fleet end-to-end — from provisioning and patching to incident recovery and cost control.
Designed an in-house security platform and applies least-privilege, patching and change-control discipline across the fleet.
Turns business needs into precise specifications and holds every release to a high bar for usability and accessibility.
Runs a multi-domain web estate with search visibility, analytics and performance treated as first-class requirements.
Ships production changes by directing AI coding agents, and builds AI into operations tooling with humans kept in control.
Moved MySQL backups to weekly full dumps plus daily binary-log backups, which ended oversized daily dumps on one server. Added off-site copies to Amazon S3, plus a daily verification digest that re-reads the whole restore chain (full dump + every binlog since) and raises a priority alert on any gap.
OpenAI and Cursor agents investigate server findings through a read-only command gate and propose a fix. A safety validator checks the plan, and nothing is applied until an admin approves it. Includes an AI key and model admin page and reusable fix playbooks.
Built "Fix All Servers" and a Non-apt Updates tool: parallel fixes, kernel clean-up with reboots, Tomcat patching and a Node.js 22 LTS upgrade path with automatic rollback. Amazon Inspector findings now show per server, already-fixed items are hidden, and every finding links to the tool that can fix it.
Rotated server-to-server SSH keys on all four servers, reissued TLS certificates with new private keys, and gave every application its own least-privilege MySQL account before rotating the database admin password.
Responded to the 2026 critical PHP advisories: patched all PHP servers, enabled Ubuntu Pro security coverage on the whole fleet, tightened PHP runtime settings and removed risky diagnostic files from public folders.
Admin root access is now localhost-only. Named accounts are restricted by source IP, with automatic lockout after five failed logins and Fail2Ban blocking repeat offenders. Remote management goes through one dedicated audit account.
Scoped IAM credentials that give the dashboard exactly what it needs: EC2 read, security-group edits and reboots, S3 upload, SES email and billing read. Moved outbound email to Amazon SES.
An external security researcher reported a reflected XSS on yagati.com. Shipped the fix through a reviewed pull request the same day and moved site keys out of the code into server-side secrets.
Launched vertexprime.in from scratch: hosting and SSL, SEO and WhatsApp link previews, per-page visit auditing with time-on-page, an events hub and Ganesh Utsav 2026 pages. Also built a live Tambola game with a mobile player view, an organizer console and 2-second sync, used by residents during the festival.
A family insurance portal backed by a database, with Google sign-in, admin screens and a public view that hides personal data. Also a family tours site with journey maps.
Rankings compare these skill areas with each other, based on production work across spoorthy.org, spoorthy.net, spoorthy.in, yagati.com and the Spoorthy AWS fleet. Last reviewed November 2026. Full career history: Executive Portfolio.
