Shakeer.Ahmad Back to portfolio

The longer version.

What I do at work, how each project came together, and the real problems I've worked through.

Work experience

M365 Engineer

Princeton IT Services, Inc., Gudivada, Andhra Pradesh

I look after Microsoft 365 for several client organisations: their accounts and sign-ins, their email, moving them between systems, and the support desk that handles their requests. I also build the automation behind that desk.

Support desk and ticketing

  • Built the mail-to-ticket system in Dynamics 365 Customer Service. The support mailbox now creates, threads and routes cases by itself, cutting manual ticket creation by ~80% and getting first replies out in under 10 minutes. How it works
  • Designed how cases are sorted (work type, category, who we're waiting on, how many people are affected), so the queue shows what needs attention first.
  • Work requests end to end: diagnose, fix or escalate, update the customer, and close.

Email (Exchange Online)

  • Run Exchange Online for 50+ users: new mailboxes, shared mailboxes, Send-As and delegate permissions, and quarantine.
  • Investigate mail that goes missing or arrives late, using message trace, inbox and transport rules, and the third-party mail security filters that sit in front of Exchange.
  • Escalate tenant-level problems to Microsoft with clear evidence and business impact. Case studies

Accounts and security (Entra ID)

  • Handle joiners and leavers: create accounts, assign licences and groups, and remove access cleanly when someone leaves.
  • Manage admin roles on least privilege, enforce MFA, and maintain Conditional Access policies.
  • Run quarterly access reviews so old access doesn't pile up.

Migrations

  • Completed 3+ tenant-to-tenant migrations covering mail, OneDrive and SharePoint, including US-to-India moves for compliance, with zero data loss.
  • Moved a 40-user company from Google Workspace to Microsoft 365 without interrupting email.
  • Plan the cutover: pre-sync, DNS changes (MX, SPF, DKIM), final delta sync and checks after the move.
  • Advise on bigger changes, such as bringing several tenants under one brand, including what each option costs in licensing.

Automation

  • Built the Employee Engagement Bot in Copilot Studio and Power Automate for automatic birthday and anniversary recognition, designed so it never reveals private details.
  • Use PowerShell for bulk admin work and quick checks across tenants.
  • Exchange Online
  • Entra ID
  • Dynamics 365
  • Power Automate
  • Copilot Studio
  • AvePoint Fly
  • PowerShell

Projects

Mail-to-Ticket System

The shared support mailbox was a general inbox. I turned it into a ticketing system in Dynamics 365 Customer Service, so every real request becomes a tracked, classified and routed case. About 80% less manual ticket work, and first response in under 10 minutes.

  1. Email arrives
  2. ARC rule filters
  3. Case created or threaded
  4. Classified
  5. Power Automate routes
  6. Agent works the case

1. Mail intake

  • Connected the shared support mailbox to a Dynamics queue with server-side synchronization for incoming and outgoing mail.
  • Set up an Automatic Record Creation (ARC) rule so a case is created whenever the support address is in To or Cc.
  • Kept the original mail in the mailbox, so nothing is lost if a rule changes.

2. Noise filtering

  • Studied what actually lands in the mailbox, then replaced broad keyword filters with precise ones. "alert" and "noreply" were blocking real requests.
  • The rule now skips monitoring-tool alerts, quarantine digests, webinars and unsubscribe/marketing mail before any case is created.
  • Tested with mail that should and shouldn't become cases, and checked each result in the ARC Activity Monitor.

3. Threading and no duplicates

  • Turned on email correlation, tracking tokens and smart matching, so replies attach to the existing case instead of opening a new one.
  • Stopped duplicate timeline entries caused by mail being tracked twice between internal users.

4. Classification model

  • Built all changes in a dedicated solution on the standard Case table, so they can be moved between environments.
  • Work Type (Incident, Service Request, Monitoring Alert, Project Work, Vendor Escalation, Security Event, Administrative/FYI, Manual Review), plus Category, Source Type, Waiting On and Impact Scope.
  • Vendor Ticket ID and Related Project fields, plus monitoring fields: alert device, alert state, first/last detected, recovery received.

5. Routing and rules

  • Three review queues (Manual Review, Monitoring Review, Project & Vendor Review) behind a routing ruleset.
  • A Power Automate flow that routes each case by Work Type, and another that keeps Status Reason in sync with Waiting On. I used a flow because business rules can't reliably set status.
  • Business rules require the right fields for vendor escalations, project work and monitoring alerts.

6. Agent experience

  • Reworked the case form into a 35/65 layout so the timeline and Copilot panel have room. Removed panels nobody used.
  • Turned on Copilot case summaries and email drafting.
  • Rebuilt the case views around Status Reason, Waiting On, Work Type and Impact Scope, with newest cases first.

7. Hardening

  • Fixed internal staff being recorded as the customer on cases.
  • Turned on auditing after a case went missing, so every change can be traced.
  • Dynamics 365
  • Power Automate
  • Exchange Online
  • Dataverse
  • Copilot

Cross-Tenant M365 Migrations

Moved organisation's mail, OneDrive and SharePoint from one Microsoft 365 tenant to another, including US-to-India moves for compliance. 3+ tenants migrated with zero data loss, using AvePoint Fly and a repeatable runbook.

  1. Assess
  2. Prepare target
  3. Pre-sync
  4. Cutover
  5. Delta sync
  6. Validate

1. Assessment and scope

  • Listed what had to move: mailboxes, OneDrive, SharePoint sites and libraries, Teams chat and any shared mailboxes.
  • Separated Teams teams (channels and files) from private chats, since migration tools handle and license them differently.
  • Confirmed data sizes and the cutover date with the client before work started.

2. Preparing the target tenant

  • Migration tools move content, not identities, so I created every user in the target tenant first and assigned licences.
  • Built a one-to-one mapping from each source account to its new target account (for example, a user at the old domain to the same user at the new one).
  • Recreated groups and shared mailboxes with matching permissions.

3. Pre-sync and cutover

  • Ran a staged pre-sync with AvePoint Fly while users kept working in the old tenant, so most data was moved before cutover day.
  • On cutover, moved the domain and updated MX, SPF, Autodiscover and DKIM so new mail landed in the target tenant.
  • Ran a final delta sync to catch anything that arrived during the switch.

4. Validation and handover

  • Checked item counts, mailbox permissions and file access, and sent test mail in both directions.
  • Watched mail flow with message trace for the first days after cutover.
  • Kept the source tenant active as a fallback until everything was confirmed.

5. Documentation

  • Wrote an internal SOP and step-by-step runbook for AvePoint cross-tenant migrations, plus a pre- and post-migration checklist, so each new migration follows the same steps.
  • AvePoint Fly
  • Exchange Online
  • OneDrive
  • SharePoint
  • DNS

Google Workspace → Microsoft 365

Migrated a 40-user company's mail, calendar, contacts and Drive files from Google Workspace to Microsoft 365, and ran the cutover on a live call with the client so email delivery never stopped.

  1. Migrate data
  2. Final delta sync
  3. DNS cutover
  4. Fix conflicts
  5. Monitor
  6. Retire Google

1. Moving the data

  • Moved Gmail, Calendar and Contacts into Exchange Online, and Drive files into OneDrive and SharePoint.
  • Ran delta syncs right up to cutover so nothing sent in the meantime was missed.

2. DNS cutover

  • Used the Microsoft 365 domain setup with the registrar's automatic DNS option to add MX, Autodiscover, SPF and DKIM records.
  • Told the client's harmless leftover records apart from the ones that actually break mail. Only MX and SPF had to be replaced; the other old records were safe to leave.

3. Fixing a blocked cutover

  • The domain health check flagged errors because five old Google MX records were still active next to Microsoft's, which would have split incoming mail between two providers.
  • The registrar refused to delete them. I traced the lock to a mail service still linked to the domain, then cleared it with the domain owner during the call.
  • Merged SPF into a single Microsoft record, since two SPF records make outgoing mail fail checks.

4. After cutover

  • Ran a final delta sync a few hours later to pick up mail that reached Gmail while DNS changes spread.
  • Kept Google Workspace running for about two weeks as a safety net, then retired it once mail flow was stable.
  • Handled follow-up requests such as changing a user's primary address while keeping the old one as an alias.
  • Exchange Online
  • OneDrive
  • DNS
  • MX / SPF / DKIM

Employee Engagement Bot

Automated birthday and work-anniversary recognition: Teams posts, branded HTML emails, and a Copilot Studio bot in Teams that employees can ask about upcoming celebrations.

  • An Employee Directory in a SharePoint list, bulk-loaded from Excel.
  • Scheduled Power Automate flows check the list every day, post in Teams and send a branded HTML email. I later redesigned the email to look more modern.
  • Published the bot to the Teams app catalog for the whole company and fixed access problems for individual users.
  • Privacy fix: with SharePoint as a knowledge source, each user needed access to the list itself, which would have exposed private employee information. It also showed source links in answers. I replaced it with a Power Automate tool flow that reads the list on its own connection and returns only the fields the bot needs.
  • Wrote clear agent instructions: a "fetching details" status message, no links or private details, no guessing, and other topics sent to IT Support.
  • Copilot Studio
  • Power Automate
  • SharePoint
  • Teams

Biometric Access & Cloud Attendance

Deployed a face and fingerprint terminal for an office of 20+ employees, then moved the whole setup from a locally hosted server to a cloud platform, so management no longer depends on one workstation staying online.

  1. Terminal setup
  2. Local server
  3. Data migration
  4. Cloud platform
  5. Mobile and reports
  • Built daytime, overnight and cross-midnight attendance schedules with different weekly offs per group. First-in and last-out logic stops stray mid-shift detections from distorting worked hours.
  • Fixed an incident where the terminal was online but events stopped reaching the server. The cause was a stale event channel held by the wrong client. A controlled restart restored live events, and the missed history was recovered from the device with no reset and no re-enrolment.
  • Migrated employees, departments and credentials to the cloud platform by reshaping the export into its import format, matching on employee ID to avoid duplicates.
  • Proved the cloud path by stopping the local services and confirming new events still arrived automatically.
  • Hikvision
  • HikCentral
  • Hik-Connect
  • ISAPI
  • PowerShell

Cloud Portfolio & CI/CD

Production static portfolio hosted on Azure Static Web Apps with GitHub Actions-based continuous deployment. Every push to the main branch automatically publishes the latest version.

  1. GitHub repository
  2. GitHub Actions
  3. Azure Static Web Apps
  4. shakeer.space
  • Azure Static Web Apps hosts the production static site.
  • GitHub Actions automatically deploys changes pushed to main.
  • Cloudflare manages DNS for the custom domain shakeer.space.
  • HTTPS and the custom domain are configured through Azure Static Web Apps.
  • Previously hosted on AWS S3/CloudFront; migrated to Azure Static Web Apps.
  • Azure Static Web Apps
  • GitHub Actions
  • Cloudflare
  • CI/CD

View on GitHub

Case studies

Real problems from day-to-day work. Each one shows the situation and what I did. Names are left out.

Tenant blocked from sending mail

SituationA Microsoft outbound block (550 5.7.705) stopped every external email from a tenant for more than six days.

  • Ruled out DNS and connector causes and reviewed the domain's mail-flow risk report.
  • When the Microsoft case stalled, escalated with business-impact evidence to the engineer, technical lead and manager.
  • Drafted a business-continuity plan for critical communication while the block was in place.
  • Exchange Online
  • Incident management
  • Escalation

Consolidating seven tenants under one brand

SituationAn organisation with seven Microsoft 365 tenants wanted everything under one company name.

  • Laid out the two real options: a Multitenant Organization (shared communication, tenants stay separate) or a full tenant-to-tenant migration (one tenant, one identity).
  • Found the licensing gap early: MTO cross-tenant sync needs Entra ID P1, and the tenants were on Business Basic.
  • Recommended a pilot group first, with MTO as a coexistence stage if they later migrate.
  • Entra ID
  • Multitenant Organization
  • Licensing