~/jesse-obymako

IT support and systems · AI workflow automation · Cloud

AI automation with a human in the lead.

More than four years on an enterprise service desk taught me to find the failure people can’t see. Now I build AI workflows and cloud tooling where the agents draft and test the work, and nothing ships under my name until I approve it.

Portrait of Jesse Obymako

Live GitHub stats

Pulled live from GitHub. Every fix includes a test.

Selected work

Proof over claims. Each card links to the code or the pull request.

llmops-infra-guide

An LLM drafts a Dockerfile, docker build validates it, and the build errors feed the next attempt. A generate-and-validate loop with documented samples for Flask, Node.js and Go.

  • LLMOps
  • Docker
  • Python
  • Node.js
  • Go
View on GitHub (llmops-infra-guide)

Merged upstream merged

career-ops #7098: Updater drift reporting

When the commit lookup failed, the updater said “up to date”. It now reports drift, and a test stubs the failed lookup.

  • Node.js
  • Reliability
  • Test added

Pull request state Read the pull request for career-ops #7098

career-ops #3711: Playwright MCP detection

Detects project-scoped Playwright MCP installs, with precedence-aware config resolution and regression tests.

  • Playwright
  • MCP
  • Config
  • Test added

Pull request state Read the pull request for career-ops #3711

sure #5038: Brandfetch logo URLs

Stored logo URLs resolve with the current client ID, so rotating the ID repairs existing logos without a data backfill.

  • Ruby on Rails
  • Integrations
  • Test added

Pull request state Read the pull request for sure #5038

ttt #1601: Indent settings for new buffers

New untitled buffers inherit the editor’s insertSpaces and tabSize settings, verified against the built binary.

  • Editor
  • Bug fix
  • Test added

Pull request state Read the pull request for ttt #1601

In review upstream

conductor #1831: MySQL unack sweep

The unack sweep stays scheduled after a failed run, so a dropped database connection no longer strands workflows until restart.

  • Java
  • MySQL
  • Workflow engine
  • Test added

Pull request state Read the pull request for conductor #1831

agent-browser #2549: Cross-origin iframe selector

frame "#selector" resolves cross-origin iframes, covered by an end-to-end test against real Chrome.

  • Browser automation
  • Chrome
  • E2E test

Pull request state Read the pull request for agent-browser #2549

Building now (private)

Private projects in progress. Happy to walk through them in an interview.

AI-assisted IT service desk

Agents handle intake, triage and first-line remediation. A fix that fails review gets up to three examined fixes before the ticket escalates to a technician, and privileged actions such as password resets need a human approver.

  • ITSM
  • AI agents
  • Approval gates

Operator console

Halt a run, preserve its session, and review the report every run produces. Any run can be handed to a person at the same step.

  • Observability
  • Human in the lead
  • Run reports
GitHub contribution graph with a snake eating the contribution squares

How I work

The operating model borrows from IT service management: clear tickets, clear owners, and a person who signs off.

How I workI brief a personal agent, a lead splits the work across a group of specialists, and results come back through the lead and the personal agent to an approval gate for me. Approved work goes back to the specialists to finish and verify. Mehuman in the lead Personal agentbriefs and presents Leadsplits the work Specialistschecked independently Finishverified Approval results for my sign-off approved work goes back to the specialists to finish

I brief a personal agent, a lead splits the work across specialists, and results come back to me for approval before anything is finished.

Brief first

Every task starts as a written brief with scope, done criteria and what is off limits.

Specialist work

Build, test, security review and writing each go to the specialist suited to them, checked independently. A summary from another agent is not evidence.

Approval gate

Pull requests, public posts and anything that spends money wait for my approval of the exact item. Approved work is then finished and verified.

Examined fixes

On failure: what broke, the likely cause, what the next attempt changes. At most three, then it halts and escalates with the evidence.

Experience

AI Systems & Automation · Self-employed

Aug 2026 – Present

  • Designed and run a supervised multi-agent AI workflow on Claude Code and Copilot CLI, among other tools, for engineering, research and audit work, with service-desk-style escalation.
  • Built approval gates: nothing pushes code, opens a pull request, posts or spends without my sign-off, and a blocked action halts with full context instead of retrying.
  • Kept AI costs low by matching each task to the least expensive model that does it at full quality.
  • Wrote specifications, runbooks and schema documentation, with automated tests that keep definitions consistent across platforms.
  • Contributed tested fixes upstream to career-ops, sure and ttt (merged) and to conductor and agent-browser (in review).

Career break: professional development and family caregiving

Mar 2022 – Aug 2026

  • Earned CompTIA A+ (Dec 2023).
  • Self-directed study in full-stack development, CI/CD pipelines, cloud infrastructure and AI agent orchestration.
  • Primary caregiver for elderly family members.
  • Earned SMAW and MIG welding certifications (Jun 2023), with working knowledge of TIG.

Service Desk Analyst · Harbor Freight Tools

Oct 2017 – Mar 2022

  • Resolved hardware, software, network and enterprise-application issues for store and corporate users by phone, chat, email and ticket, within response-time targets.
  • Found a recurring POS failure pattern and built an automated fix, cutting service desk call volume by about 30%.
  • Diagnosed network outages across a nationwide store network, restoring locations that could not operate without their systems.
  • Supported the move from on-premises infrastructure to a Microsoft Azure hybrid environment and the migration to ServiceNow (ticketing, workflows, back-end configuration).
  • Provided priority support to retail VPs; researched account history in Salesforce and SQL; supported the in-house credit card rollout.
  • Escalated defects to development, networking and security teams through Jira and worked with vendors on replacement hardware.

Supervisor / Key Holder · Jersey Mike’s Subs

Mar 2016 – Oct 2017

  • Ran opening and closing: register reconciliation, cash handling, deposits and inventory.
  • Supervised and coached staff and resolved customer issues on the spot.

Skills

Support & ITSM

Help desk and service desk, high-volume ticket queues, remote support, escalation management, root-cause analysis, knowledge-base writing

Systems, Cloud & Networking

Windows, Windows Server, Active Directory, Microsoft Azure (hybrid), AWS, VMware, Docker, Cisco routers and switches, network diagnostics, telephony

Code & Automation

Python, PowerShell, Bash, SQL, Node.js, C, Java, Git and GitHub, Playwright, CI/CD

AI Workflows & Tools

Claude Code and Copilot CLI, among other tools; approval-gated agent workflows; LLM generate-and-validate loops; ServiceNow, Jira, Salesforce, Teams, Slack, Zoom

CompTIA A+ · Earned December 2023

Let’s talk

Open to AI automation, cloud and technical support roles, in the Los Angeles / Long Beach area or remote.