Bernardo Gallegos Vallejo
7+ years of experience supporting enterprise infrastructure. From AWS data centers to AI/ML workloads at Cirrascale, I bridge hardware and software to create reliable, scalable solutions.
From Hardware to Code, Building the Future
My journey began in Amazon's warehouses, supporting network infrastructure that keeps operations running for thousands of users. I learned that behind every seamless experience lies carefully orchestrated hardware and the people who understand both its capabilities and limitations.
Moving to AWS data centers opened my eyes to enterprise-scale infrastructure. Managing 20,000+ devices taught me that reliability isn't just about robust hardware—it's about predictive monitoring, automated responses, and building systems that recover gracefully from failures. Contributing to 5MW data center expansions showed me how physical planning and logical architecture must work in harmony.
Now at Cirrascale Cloud Services, I'm at the intersection of traditional data center operations and cutting-edge AI infrastructure. Developing automation scripts for 5,000+ servers and supporting NVIDIA GPU deployments means every line of code I write directly impacts the compute power driving tomorrow's breakthroughs.
What drives me is turning complex infrastructure challenges into elegant, automated solutions. Whether it's creating monitoring systems that prevent failures before they happen, or building the network backbone that connects AI clusters, I find purpose in making the invisible foundation of technology both reliable and scalable.
"I believe the most impactful infrastructure engineers understand both the physics of hardware and the power of automation. In our AI-driven future, the data centers and networks we build today are the foundation for innovations we can barely imagine tomorrow."
Ready to build the future of infrastructure together?
Technical Skills
A comprehensive toolkit built through years of hands-on experience managing enterprise infrastructure and GPU clusters.
Data Center Technologies
Network & Systems
Infrastructure Management
Technical Philosophy
Infrastructure as Code
Everything should be version-controlled, reproducible, and automated.
Monitoring First
You can't improve what you can't measure. Observability is crucial.
Hardware + Software
Understanding both physical infrastructure and intelligent automation.
Experience
A track record of delivering reliable infrastructure solutions and leading automation initiatives in enterprise environments.
Network Operations Engineer
Key Achievements:
- ▸Developed Python and Bash scripts for BMC monitoring across 5,000+ servers, automating health checks and predictive failure analysis
- ▸Configured and maintained Cisco and Arista switches for computing clusters, implementing VLAN segmentation and routing protocols
- ▸Assisted with NVIDIA GPU deployments for AI/ML workloads, including driver updates and thermal monitoring
- ▸Built monitoring solutions integrating IPMI, SNMP, and custom APIs to track power and performance metrics
- ▸Created comprehensive documentation including runbooks, network diagrams, and SOPs in ServiceNow
Technologies:
Data Center Engineer
Key Achievements:
- ▸Participated in new data center construction including rack layout planning, power distribution design, and network topology for 5MW expansion
- ▸Supported data center operations for 20,000+ devices including server deployment, network infrastructure, and power/cooling systems
- ▸Assisted with AI/ML compute cluster design and deployment including rack placement, InfiniBand topology, and GPU server configurations
- ▸Maintained PXE boot infrastructure for automated OS deployment and implemented ZTP for network equipment
- ▸Designed and installed fiber optic and copper cabling for new builds including pathway planning and cable tray installation
- ▸Maintained 99.9% uptime through proactive monitoring and rapid incident response
Technologies:
IT Technician
Key Achievements:
- ▸Provided first-level support for campus network infrastructure serving 3,000+ users
- ▸Tracked and maintained inventory of network equipment using asset management systems
- ▸Identified network bottlenecks and implemented fixes to improve warehouse system response times
- ▸Coordinated with vendors for RMA processing and equipment orders
Technologies:
Projects
Infrastructure automation and optimization projects that demonstrate practical solutions to real-world challenges.
BMC Monitoring Automation for 5,000+ Servers
Python and Bash automation suite for comprehensive server health monitoring and predictive failure analysis
Challenge:
Manual server health checks across thousands of servers were time-consuming and reactive, leading to unexpected failures and extended downtime that impacted critical AI/ML workloads.
Solution:
Developed comprehensive Python scripts integrating IPMI, SNMP, and custom APIs to automate BMC monitoring. Created predictive algorithms analyzing temperature trends, power consumption, and hardware sensor data to identify potential failures before they occur.
Automated health checks across 5,000+ servers, reduced manual monitoring time by 90%, and implemented predictive maintenance that prevented dozens of critical failures. System now provides 24/7 autonomous monitoring with intelligent alerting.
Technologies:
AI Infrastructure Network Architecture
Network design and implementation for high-performance AI/ML compute clusters
Challenge:
Supporting NVIDIA GPU deployments required specialized network configurations with VLAN segmentation, optimized routing protocols, and thermal monitoring to ensure maximum performance for distributed AI workloads.
Solution:
Configured Cisco and Arista switches for computing clusters, implementing strategic VLAN segmentation and routing protocols. Designed network topology optimized for GPU-to-GPU communication and integrated thermal monitoring systems for cluster health management.
Successfully deployed network infrastructure supporting multiple AI/ML clusters, achieved optimal GPU communication performance, and established monitoring frameworks that ensure consistent thermal management and network reliability.
Technologies:
5MW Data Center Expansion Project
Comprehensive infrastructure planning and implementation for enterprise-scale data center expansion
Challenge:
Contributing to a 5MW data center expansion required coordinating rack layout planning, power distribution design, network topology implementation, and fiber optic cabling while maintaining zero safety incidents and meeting aggressive deadlines.
Solution:
Participated in full greenfield data center buildout including rack placement optimization, power distribution planning, InfiniBand network topology design, and structured cabling installation. Coordinated with construction teams and vendors to ensure on-time delivery.
Contributed to successful on-time delivery of 5MW expansion with zero safety incidents, supporting deployment of AI/ML compute infrastructure and establishing scalable foundation for future growth.
Technologies:
PXE Boot Infrastructure & Zero Touch Provisioning
Automated deployment system for rapid server provisioning and network equipment configuration
Challenge:
Manual server provisioning and network device configuration was creating deployment bottlenecks, taking hours per device and increasing the risk of configuration errors in large-scale deployments.
Solution:
Maintained and optimized PXE boot infrastructure for automated OS deployment and implemented Zero Touch Provisioning (ZTP) for network equipment using vendor tools and Python scripts. Created standardized deployment workflows integrated with asset management systems.
Dramatically reduced deployment time from hours to minutes per device, eliminated configuration errors through automation, and established scalable provisioning processes supporting rapid infrastructure scaling.
Technologies:
Campus Network Performance Optimization
Network troubleshooting and optimization tools for enterprise campus infrastructure
Challenge:
Network bottlenecks were impacting warehouse system performance for 3,000+ users, causing delays in operations and reduced productivity across multiple facilities.
Solution:
Developed systematic approach to identify network bottlenecks using performance analysis tools and implemented targeted fixes to improve system response times. Created monitoring frameworks to prevent future performance degradation.
Significantly improved warehouse system response times, eliminated network bottlenecks affecting thousands of users, and established proactive monitoring to maintain optimal performance standards.
Technologies:
Documentation & Knowledge Management System
Comprehensive documentation platform for infrastructure operations and maintenance procedures
Challenge:
Lack of standardized documentation was leading to inconsistent procedures, extended troubleshooting times, and knowledge gaps when team members were unavailable, impacting overall operational efficiency.
Solution:
Created comprehensive runbooks, network diagrams, and standard operating procedures in ServiceNow. Developed training materials and knowledge base that reduced new technician onboarding time and standardized operational procedures across teams.
Reduced new technician onboarding time by 50%, standardized operational procedures, and created sustainable knowledge management system that improved overall team efficiency and reduced troubleshooting time.
Technologies:
Interested in Collaborating?
I'm always excited to work on new infrastructure challenges and automation projects. Let's discuss how we can build something amazing together.
Ready to Build Reliable Infrastructure?
Whether you need expertise in data center operations, network infrastructure, or automation solutions for enterprise-scale environments, let's connect and build something solid together.
Send a Message
Tell me about your infrastructure challenges
Availability
Specializing in:
- • GPU cluster management & optimization
- • Infrastructure automation & IaC
- • Predictive maintenance systems
- • Data center operations
- • High-performance computing