I keep the technology people depend on working — service desk to directory to documentation — and I automate the parts that should never have been manual. Five plus years across two of New York's largest institutions, with the working notes to show for it.
The day job spans support, administration, process, and automation. These are the areas where I do the work rather than just know the terminology.
Frontline and escalated support for staff in high-demand institutional environments — Windows endpoints, peripherals, access problems, and the long-tail issues that never match a script. Structured triage first, guesswork never.
Account lifecycle, group membership, licensing, mailbox and Teams administration, MFA enforcement, and least-privilege access reviews in Active Directory. Every change traceable to a request and a reason.
ITIL 4 practices applied to the ordinary stuff: incident and request handling, queue hygiene, knowledge articles people actually reuse, and handoffs that survive a shift change. Documentation is deliverable, not afterthought.
Python and shell automation for the repetitive work, plus production AI workflows with real guardrails — credential isolation, validation gates, retry logic, and state you can audit after the fact.
Labs and projects I run outside of work to keep sharp. Each one is documented the way I would hand it to a colleague: what the problem was, what I did, and what the evidence says.
An autonomous multi-agent pipeline that researches a brand, writes listings, generates design mockups, and publishes to a marketplace over OAuth — with credential scanning, validation gates, exponential backoff, and SQLite state tracking, deployed on a VPS under systemd.
Windows authentication logs ingested into Splunk, SPL queries written to group failed logons by source and account, thresholds tuned to cut noise, and the whole triage path written up so the next person can repeat it.
A phishing-to-malware scenario worked end to end: user report, host indicator review, containment decisions, and an incident summary written for people who were not in the room.
Access-request, recovery, and MFA scenarios documented with the decision points that matter — who approves, what proves identity, what least privilege actually resolves to, and what gets recorded.
A stuck laptop and a failing automation get the same treatment. It is the reason the write-ups look alike, and the reason the fixes hold.
Establish what is actually broken, for whom, since when, and what changed. Reproduce it before touching anything.
Go to the logs, the directory, the config. Separate signal from noise and write down what the evidence supports.
Apply the smallest change that fixes the cause, verify it with the user or the data, and note what was ruled out.
Leave a record someone else can act on — knowledge article, runbook step, or script — so the next occurrence is shorter.
I have spent five plus years in enterprise IT at institutions where downtime is not an inconvenience — Columbia University and NYC Health + Hospitals — supporting the infrastructure and the people who rely on it at scale.
My work sits where support, administration, and process meet: keeping endpoints and accounts healthy in Microsoft 365 and Active Directory, running requests and incidents through ITIL 4 practices, and increasingly replacing repetitive work with automation I can audit.
Certified in CompTIA Security+, and ITIL 4 Foundation completed in 2023. The labs and projects here are how I build depth on purpose rather than by accident.
Technology should simplify complexity, strengthen security, and create repeatable outcomes.
ITIL 4 Foundation in April 2023, CompTIA Security+ in May 2025. Each has its own page with the credential details.
Baseline security competency across threat analysis, architecture, network security, identity management, operations, governance, and risk.
Service value chain, the four dimensions model, guiding principles, and continual improvement practices as applied in enterprise IT.
