5 Years of Operational Instinct.
18 Months of Applied AI Engineering.

Spent 18 months learning AI engineering by building real production systems that bridge enterprise IT operations knowledge with the modern AI stack. Every system ships with security, observability, and real users.

• Built a live wellness platform (WellnessInYou) using Next.js 15, Expo React Native, CrewAI multi-agent pipelines, and Supabase : with hardware-encrypted AES-256-GCM credential storage and PII-scrubbing telemetry.
• Designed a 6-agent SDLC execution graph (Agentic SDLC) using a custom Kahn's algorithm DAG scheduler in Node.js for parallel pipeline execution across Planner, Implementer, Reviewer, Test, Doc, and Summarizer agents.
• Built Founder Growth OS: a 22-page React CRM with Firecrawl lead enrichment, Hinglish DM generation via Supabase Edge Functions, Instagram comment-to-DM webhook triggers, and multi-channel content publishing.
• Engineered reusable production scaffolds across projects: circuit breakers, Redis task queues, HMAC webhook validation, multi-provider LLM routing with automatic failover.
• Extracted and published a suite of 14 open-source repositories containing real, production-ready code (rather than mock learning artifacts) directly from live commercial systems: including the core asynchronous infrastructure of AgentKernel, the complete Agentic SaaS Boilerplate, and core engines for failure memory (Agent-Scars), codebase dependency mapping (Agent-Context), and multi-provider routing (Agent-Routing).

↑ Promoted from SAP Support Consultant

Sole L2 SAP support owner for enterprise users across global clients — end-to-end ownership of ticket lifecycle, workflow design, stakeholder communication, and continuous improvement.

• 98% SLA compliance across a sustained queue of 350+ tickets per week : structured triage, clear escalation paths, and zero ticket backlog drift.
• 95% MTTR improvement over tenure : embedded recurring issue fixes directly into runbooks so patterns were eliminated at source, not resolved repeatedly.
• Owned all P1/P2/P3 incident management for SAP operations : root-cause analysis, proactive monitoring via Graylog dashboards, and coordination across users, infra, dev, and vendor teams.
• Redesigned ServiceNow ticket workflows to improve request routing consistency and reduce handoff delays across support tiers.
• Created RCA documentation and ran team reviews to prevent recurring issues from re-entering the queue.
• Built runbooks and monitoring practices that caught system issues before they reached business impact.

Joined as a 3rd-party L1 support agent and earned four progressive promotions within the same engagement: L1 Agent → SME → Team Lead → Acting Floor Manager. Grew from handling tickets to owning operations, training pipelines, and new vertical launches across a 250+ seat support floor.

Leadership and Scale

• Onboarded and trained 250+ support agents : designed structured knowledge bases, ran live training sessions and webinars, and provided hands-on floor coaching through go-live phases across multiple programme intakes.
• Acted as Floor Manager during high-traffic escalation windows : owned live queue prioritization, real-time P1/P2 SLA coordination with client stakeholders, and incident triage decisions for the full floor.
• Built a culture of ownership and documentation on the floor : established shared runbooks, team SOPs, and recurring office-hours Q&A sessions that reduced escalation noise and improved first-contact resolution.

New Vertical Launch

• Led a new chat and call support vertical from zero to live in 6 months : scoped the programme, designed the support workflows, built the onboarding documentation, and deployed a fully trained team to production on schedule.
• Defined channel-specific escalation paths and SLA targets for the new vertical, embedding quality checkpoints that ensured consistent client experience from day one.

Systems and Automation

• Managed SCCM patch deployment cycles and authored compliance-minded operational runbooks covering vulnerability checks, patch cadence, and incident-response handoffs.
• Used SnapLogic (iPaaS) to build integration workflows that automated manual handoff steps between ServiceNow modules : reusable playbooks adopted across the programme.
• Converted recurring user friction points into structured product feedback loops reviewed by client engineering and product teams, improving platform-wide content coverage.