Nick Mccollum Email & Phone Number

Decatur, Al, Us

Decatur, Al, Us

Fremont, Ca, Us

* Primary responsibility of developing the architecture for our AI infrastructure platform and ensuring it meets the needs of our clients* Currently leading a team of engineers in designing, building, and testing scalable infrastructure for AI research systems* Collaborating with product managers to define product roadmaps and feature sets based on customer feedback* Conducting research on emerging technologies and trends in artificial intelligence to identify potential opportunities for innovation* Lead a team responsible for deploying and managing the world's largest AI GPU cluster at the time (2021-2023), overseeing all aspects related to the installation and configuration of 16k A100 GPU’s plus associated storage devices while simultaneously overseeing maintenance activities throughout its lifecycle* Oversaw and contributed automation code to manage thousands of nodes, 24000+ GPU’s and multiple Ethernet and InfiniBand networks while maintaining >95% uptime* Developed code to automate health monitoring of large InfiniBand networks* Developed code to automate health checking and qualification of GPU hardware* Subject matter expert for large scale Slurm clusters* Develop and document best practices so others can benefit from lessons learned

Fremont, Ca, Us

Falls Church, Va, Us

- Architect, build and maintain a multi-thousand core heterogeneous dense memory cluster for Alabama Supercomputer Authority.- Manage an SGI Ultraviolet 2000 with 256 Sandy Bridge cores- Deployed and maintained a 180TB DDN GPFS parallel file storage solution.- Developed and maintain nightly backups to on-site and off-site locations utilizing parallel rsync combined with ZFS snapshots- Monitor cluster performance as well as network performance to ensure optimal utilization- Maintain multiple layers of network security- Developed and continually evolve a self-healing script that automatically detects problems and fixes itself before users notice a problem- Moved all services from physical nodes to redundant highly-available virtual machines.- Migrated from Torque/Moab to SLURM queue system.- Manage firewalls, migrating from Blacklists to Whitelists.- Wrote software to scrape cgroup metrics, write the data to InfluxDB and provide users job graphs via Grafana.- Automated node provisioning with PXE, Kickstart and Salt.- Wrote a job submission plugin for SLURM to help notify users if they submitted their jobs with invalid parameters.- Implemented a Docker solution for particularly troubling software.- Work with vendors to get quotes for hardware upgrades.- Developed wrapper scripts for many SLURM commands to output data to users.

Global, Us

• Manage numerous client e-mail servers for Alabama Supercomputer Authority (ASA) to include postfix, dovecot, openLDAP, courier-IMAP and webmail administration.• Monitor and maintain multiple geographically diverse backup servers to ensure that 200+ servers are backed up regularly in case of a disaster.• Engineer new solutions for upgrading services to compete with ASA’s competitors.• Develop solutions to automate day-to-day tasks in order to increase efficiency. • Setup and deploy virtual and physical servers for new clients and upgrade existing clients to new services.• Assist technical coordinators at local school systems throughout Alabama with various issues.• Research, deploy, and test new technologies to see if they can be integrated into ASA services.• Manage Domain Name System (DNS) for ASA clients and update records for clients in a timely manner.• Monitor and repair servers to include hard drive and power supply failures on both ASA managed servers and locally hosted client owned servers at the ASA Data Center (DC).• Proactively search for service problems before clients experience a service interruption. • Maintain firewalls to ensure that hackers cannot access ASA assets.• Update software on all Linux Systems Administration (LSA) managed servers to prevent attackers from using known exploits.• Monitor automated logwatch e-mails to ensure that servers are not being exploited.• Quickly and accurately lock compromised accounts on client servers and mitigate the effects of the compromised account to ensure clients do not get blacklisted.

• Provided worldwide onsite support for AFCENT and numerous USAF Air Control Squadrons including Kandahar Air Base Afghanistan, Al Udeid Air Base Qatar, Al Dhafra Air Base United Arab Emirates.• Performed preventive maintenance, troubleshooting, repair, and assessment of damages/failures of all systems of the Battlefield Command and Control Center (BC3) to include the following equipment suppliers: Ultra, Harris, Raytheon, VIASAT, SafeNet, General Dynamics, Cisco, L3, and Sensis electronics systems and computers, networking internal to the shelter, and external connections/networking on the military networks required for the operation of the BC3.• Performed installation of Mobile Training Suite at deployed locations to provide spin up training to USAF Air Control Squadrons preparing to deploy to CENTCOM Area of Operations.• Maintained, repaired, and installed various types of electronic equipment and related devices such as electronic transmitting and receiving equipment, servers, measuring and controlling equipment, satellite communications equipment and associated electrical and electronic devices in a field environment.• Provided System Administration of Windows and Solaris servers and workstations.• Provided communications support of HF, UHF, VHF & MIDS radios to include troubleshooting, repair, and installation for air traffic control and tactical air control operations.• Provides Link-11 & Link-16 data link support on the Joint Range Extension gateway & Air Defense Systems Integrator for the Air Force Interface Control Technicians.• Provided critical analysis for USAF Operators on the Tactical Display Framework in manipulation of Air Tasking Orders & Airspace Control Orders.• Researched and ordered repair and replacement parts and components.• Properly safeguarded & handled Controlled Cryptographic Item (CCI) equipment.• Open/Track/Close work orders for field equipment.

Randolph Afb, Tx, Us

• Operating, troubleshooting, and maintaining satellite transmission systems: TSC-179, USC-60, ICEv3.• Operated the following military radios: PRC-152, PRC-117, Motorola Astro 5000 series.• Apply working knowledge of troubleshooting links and signal flow.• Installing and dismantling UHF and VHF antennas and large aperture antennas for SATCOM.

Randolph Afb, Tx, Us

• Assisted with removal and installation of vehicle mounted radios: PRC-152 w/ VAU-110, Advanced Singars Improvement Program (ASIP) radios.• Provided audio/visual support to 332nd Air Expeditionary Wing briefings ensuring personnel received critical Joint Base Balad operations information.• Maintained 1,800 Motorola LMR assets affording reliable, secure communications of Joint Base Balad.• Assisted in Amnesty Day, sorting assets for disposition at Joint Base Balad, Iraq.• Oversaw both radio and COMSEC training programs, developing presentations and training 600+ new personnel.• Ensured 332nd and 532nd Expeditionary Security Forces Group combat capability; maintained 1,000+ equipment items in 56 tactical vehicles.• Identified and inventoried unused communications equipment and assisted in redistributing over $2 million worth of assets into Air Force Central Command's inventory.

• Assembled hardware for both workstations and servers and delivered and installed systems on-site.• Maintained and troubleshot small networks consisting of domain controllers, exchange servers, file servers and ten or more workstations.• Identified quarantined and disinfected workstations and servers of malware and other malicious code.• Assisted clients with minor computer troubles via phone or remote desktop.• Developed slide shows and teaching classes for clients interested in learning more about safe computing practices and avoiding malicious software on the internet.• Set up automated back-up of client servers and workstation; retrieved the data for storage off site in case of fire or natural disaster.• Developed a system to lock down and remove capabilities from client workstations to ensure workstations were effectively used for only business needs.• Assisted customers and entered their computer into an automated tracking system.• Backed up client workstations for safe and secure temporary storage.• Setup an automated system so multiple computers could be checked for malicious software simultaneously.• Identified quarantined and disinfected client PCs.• Assisted customers over the phone with minor computer problems and troubleshooting procedures.• Assembled custom PCs to client specifications.• Troubleshot hardware and software issues on both Windows and Macintosh based PCs.

Education record

Education record

Education record

Education record