Developed and delivered trusted AI solutions, specializing in NLP empowering organizations to make informed real-time decisions with advanced AI-driven insights in high-stakes environments.Automated ML workflow deployments using GitHub Actions and transitioned production ML deliverables from manual Kubernetes deployments to ArgoCD automated workflows.Led the client’s ML model delivery lifecycle, from pre-contract negotiations to implementation, ensuring adherence to milestones and compliance with strict confidentiality requirements.Consolidated 43% of development infrastructure and optimized 50% of redundant ML infrastructure by parallelizing workloads on GPU instances leading to an 80% decrease in build and performance times.Developed an IaC platform and implemented GitOps automation, achieving a 93% reduction in fulfillment times for sales MVPs and ML research, enabling self-service provisioning for engineers.

Worked as the main DevOps engineer as the number of engineers increased from 75 to 200. Initially provided the DevOps engineering for the space operations team as it launched 4 high-profile satellites. Joined the platform operations team as a founding member, built its DevOps infrastructure, and helped the engineers maximize efficiency. Enabled the formerly siloed teams to optimally work together by creating unified processes, aligning their goals, and fulfilling their infrastructure requests. Decreased fulfillment times of infrastructure requests from >2 weeks to 10 minutes by designing and developing a system that allowed rapid deployment of cloud infrastructure, creating a testing sandbox environment, developing a universal CI pipeline, optimizing containerization, and building new Infrastructure as Code (Terraform / Terragrunt).Eliminated ˜50% of the time for developers to start projects by creating standard processes for code delivery and introducing templates (projects initially had their own processes). Reduced build times by ˜50%–˜80% by creating a unified toolkit which removed 8K lines of repeated code, containerizing processes, and working with the ground segment team to create updated versions of GitLab modules. Subsequently led an org-wide deployment for Kubernetes-based GitLab runners. Played a key role in establishing the platform team’s new engineering practices, and overhauled its cloud infrastructure for performance and security.Used infrastructure as code to enable the platform team to deliver code with >30% more reliability.Enabled the platform team to conduct cloud development as if it were local to users’ machines by building a tools cluster that provided a central location to deploy and test new open source and internal tooling.Prevented system-wide outages by creating a redundant container registry for production system. Found the need for new software tools, evaluated vendors, and managed deployments.

Took charge of industry-leading cloud governance automation company’s main software stack when core software people were leaving. Increased its performance and usability by consolidating formerly fragmented technologies and enhancing DevOps. Developed and released first-of-kind CI/CD pipeline; heavy focus on Terraform and Infrastructure as Code. Made 300 commits of code, including infrastructure, repositories, configurations, templates, and tests. Helped integrate the stack with AWS’s CloudFormation and Azure, and optimized monitoring for compute. Decreased cloud costs $80K per year by optimizing computing resources and centralizing cloud management.Reduced errors and bugs at least 50% by expanding the testing from end to end. Added more frontend and unit testing, and improved data monitoring. Increased release cycle efficiency by establishing weekly Agile standups (for the backend and frontend) and working as the company’s first scrum master. Also worked as the daily board master for Kanban; led daily stand-ups for DevOps. Managed resource allocation, and tracked tech debt. Tracked the most critical production issues for a Fortune 50 client ( including alerts and monitoring), and developed on-call responsiveness.Managed the relationship with company’s largest customer. Enhanced its support portal, and worked as the main bridge between client and in-house engineers.Proactively identified issues before they could have reached clients by using distributed tracking.

Helped move >100K Google employees’ work to their homes for COVID-19 safety. Maintained communications, troubleshot their development and engineering environments, and enabled internal tools to work externally. Created dashboard that revealed decreases in product quality. Implemented it for vendors, and built it into a standardized model for user data. Developed combination of SQL, GoogleSQL, and Regexp to build the desired data sets for comparing models (with some queries requiring all 3 in conjunction). Decreased bugs in products by analyzing bug queues in dashboard and data tools, and by identifying the top 10 bugs. Led meetings to find bugs’ scopes, provided testing and QA of solutions, and worked with engineers to implement them.Transformed a dashboard from performance based to showing how new features are used; created tooling suite that provides analytics. Added breakdowns by model and device.Helped empower teams to optimize releases and customer success by leading IT infrastructure and tooling improvements.

Removed bottlenecks and improved partners’ experiences by improving the vendor security model. Overcame technical, operational, and legal OKR blockers through tooling, infrastructure improvements, policy drafting, and vendor implementations. Shortened release times by creating calendar that enables forecasting of product performance. Developed complex calculator system that compares quality data on speech model according to usage history, and which discerns release times (300 engineers and 80 program/project managers adopted it). Further decreased release times by leading automation of processes in the new calendar.Cut 800 hours from data collection by developing automated Python script that provides graphs to directors. Used machine learning models to detect user sentiment, visually mapped the journeys, and found pain points. Increased staffing efficiency for global business intelligence 20% by developing infrastructure for employee allocation. Developed sub-roles and trained staff for them. As the sole designer, developed mouse-to-keyboard capabilities for Google Assistant which work on the web. Filed for U.S. patent. Co-designed and co-developed an audio layer for a smart car horn that reduces noise pollution by targeting people. Updated the documentation for the team that develops the OKG for Assistant (the documentation was 5 years behind despite 2 recent tech stack changes). Met with dozens of engineers to contextualize and map out their processes; modernized the content for bleeding-edge ML build/release cycles. Helped design and rapidly prototype a new way for smart home devices to pair with each other. Reduced engineers’ down time by managing design sprints for Google tooling—removed several multiday roadblocks. Helped create MVPs.

Increased business from HP and Canon for media-tracking web applications (Printerpoint) and print-tracking Windows application server (Argos) 100% by managing the accounts and leading implementation, enhancement, and support projects. Promoted two times. Managed projects for implementing print tracking software. Worked with SQL databases, and troubleshot on site. Enabled both clients to recover operating costs, and trained them to use the new solutions. Led Kodak’s conversion to the web solution. Replaced integrations at its sites from California to India. Decreased many of its process times and saved it time and money on asset tracking by managing project to move to the cloud. Worked as the QA engineer for the first web product (used AWS). Tracked bugs and gave advice on resolving them. Troubleshot multi-site networking issues (related to SQL, .NET, SNMP, and TCP/IP) and resolved firmware, connectivity, and logging issues. Worked as sales engineer. Helped close deals by proposing tech implementations and working as the technical point of contact. As IT manager, optimized internal systems and processes.

Conceived and directed development of API list integration solution with many services (e.g., conferencing and streaming) to fundamentally redefine how people use web applications and services (currently in strategic discussions with Google’s management). Established DevOps capability to build tools and infrastructures. Recruited the team, and provided the product development and customer success strategies. Led implementation of the tech stack, including a pioneering web framework and an OS in the website, which enabled opening of other sites to allow distributed mobile app users to open a common page and collaborate. Established and led data-driven development. Enabled integration by accessing within the website. Created functional mocks and analyzed uses. Instituted Agile processes, and tracked stories with Jira. Conducted A/B testing on users’ needs, and established the data analytics for them.

Built out the DevOps capability at company that provided technical support to Nest, Airbnb, Stripe, and Robinhood. Decreased times to respond to and resolve issues by improving IT processes. Developed the infrastructure for centralizing management and technologies. Enabled Nest support employees to increase 3x by building robust processes. Optimized the site’s security; prevented several ransomware attacks. Allowed 100% increase in business from Robinhood by leading development of its support environment, including the tech infrastructure and business processes.Connected the U.S. site’s IT (including server rooms and stacks) with HQ in Ireland. Met all legal and contracting needs. Designed and proposed system for running network operations control centers, and built out internal facility infrastructures. Also managed data security and IT security.

• Performed outsourced troubleshooting for airports, architecture & legal firms• Managing Active Directory accounts and permissions• Configuring and troubleshooting Exchange and Office 365 mail servers• Network troubleshooting and optimization• Data backup solutions and basic server maintenance • Detailed documentation of work performed via ticketing program

• Maintained client communication to enforce company rapport• Shortened average repair turn time to one of top 25 locations nationally• Managing hardware and software repairs on as many as 60 client computers at a time• Removing spyware, malware, rootkits, and bloatware from compromised computers• Restoring OS integrity, permissions, damaged services, and missing drivers