# Jaco le Roux > Senior Staff Engineer @ Crusoe.ai Location: San Francisco, California, United States Profile: https://flows.cv/jaco Hello! I am an experienced software engineer that has been in the industry for 15+ years. I’ve been in a wide variety of roles and environments, including early teams like AWS EC2 that experienced rapid growth and enormous scale in short periods of time, as well as management and other technical leadership roles at Twitch and Stedi. I excel at great communication, simplifying complex large-scale systems, and building scalable backend and infrastructure systems. Over the course of my career, I have demonstrated the ability for both depth and breadth. I enjoy collaborating with other engineers and leaders to facilitate making thoughtful forward-looking decisions, executing on large-scale projects, and reducing risk wherever possible. Some engineering highlights include: • 16 years in Software Engineering. • Specialization in architecting systems on Amazon Web Services (AWS). • Spent the majority of my career working with distributed backend systems and service-oriented architecture. • Plenty of experience building, maintaining and operating high-volume APIs and highly available services, including monitoring, alarming and continuous deployment. • I like to apply best practices where it makes sense to do so, such as good documentation, continuous integration testing, infrastructure as code. • Go is my first choice for 90% of backend systems, and I have a fair amount of experience in it. I have spent a good portion of my career working in Java, Ruby and TypeScript. Exposure to other languages such as C, Python, JavaScript, Rust, etc. is typically in smaller projects. • Experience building, mentoring, managing, and promoting a team of 8 engineers, communicating and managing multiple concurrent projects across groups, and helping with product development and project scheduling. • Lots of experience using and owning SQL databases (MySQL, Postgres), as well as NoSQL databases (AWS DynamoDB). • Deploying and running systems on Kubernetes, and deploying a Kubernetes platform itself (using Talos), experience deploying many applications in the CNCF. • Some experience with frontend apps (JavaScript, TypeScript, React). • Can operate comfortably at both high-level and low-level detail, from inter-service architecture to poking around in the proc filesystem • Highly autonomous but also a strong team member with a high degree of ownership and the ability to plan and drive projects to completion. ## Work Experience ### Sr. Staff Software Engineer @ Crusoe Jan 2024 – Present | San Francisco, CA * Lead engineer for Crusoe’s on-prem Kubernetes platform across 5+ production regions and brought the platform from 0 to 1. Owned end-to-end design and delivery from architecture RFCs through production readiness across networking, security, datacenter bootstrapping, and multi-cluster operations. * Authored and drove alignment on core architecture decisions: Talos-based node OS/HA posture, multi-region/multi-cluster replication model, Vault-backed secrets and PKI, BGP-based routing and service advertisement, node provisioning/bootstrapping, RBAC across all surfaces (Talos API, Kube API, ArgoCD, Vault), and standardized rollout/upgrade/rollback procedures for machine configuration as well as application deployment via ArgoCD. Other aspects include deploying and working with CSIs such as OpenEBS, CNI such as Multus/Cilium, Postgres Operators, etc. * Built Kubernetes-independent routing primitives to reduce outage and reboot risk: FRR as a Talos system extension, BFD-based fast failure detection, resilient service advertisement/failover behavior, and safeguards that gate VIP advertisement on Kubernetes API health. * Partnered closely with NetEng, SiteOps, Security, and internal customer teams; facilitated stakeholder alignment, maintained roadmap, and mentored engineers across teams. * Standardized operational practices on the Cloud Compute team: improved on-call and other operational processes, and implemented the spread placement topology feature. ### Staff Engineer & Team Lead @ Stedi Jan 2020 – Jan 2023 | San Francisco, California, United States Joined and led a variety of teams and efforts during my time at this startup. 1. I was part of the Transactions team, responsible for building a secure multi-tenanted transaction services between businesses on the platform. 2. I lead a Database team, responsible for delivering an experimental Ledger database as well as Key-Value store that formed part of the Stedi platform. 3. I joined the Substrate team, responsible for all AWS organization management, Single Sign-on (SSO) systems, Identity Management, and other IT systems required by Stedi. 4. I joined the Cloud team, responsible for wrapping AWS services as useful platform modules in a single-tenant-per-AWS-account system that spanned thousands of AWS accounts. Notable work included: * Built various features and provided support for the Transactions service, a serverless B2B multi-tenanted system that created and recorded transactions between businesses. * Helped implement a mechanism for consuming and parsing the X12 standard into a machine-readable format. * Designed and delivered Stash: an easy-to-use multi-tenant key-value store that wrapped DynamoDB across thousands of AWS accounts to ensure account isolation. * Significantly reduced complexity and entanglement of a monolithic AWS organization management service (dubbed "the Substrate pipeline"). * Delivered webhook functionality for incoming X12 documents. Other notable contributions: * Introduced company-wide team update emails that other teams at the company also adopted. * Introduced "Requirement Docs" that we used to get better at defining requirements and thinking about our projects ahead of time. ### Sr. Software Engineer / Tech Lead @ Twitch Jan 2019 – Jan 2020 | San Francisco Bay Area While continuing to provide management support for the VOD Infra team, I transitioned to an engineering-focused role. This time was primarily focused on one project that was significant in scope. I lead the product design, architecture, cross-team coordination of the VOD side of the new Amazon Interactive Video Service (IVS). My team and I built multiple proof-of-concept services for internal AWS customers such as IMDB and Prime Video. I was responsible for documenting and exploring all product requirements, as well as driving the project forward. There was a heavy focus on implementing the product securely to meet AWS security standards, designing the system with multi-region resilience as well as network capacity planning. ### Software Development Manager III @ Twitch Jan 2017 – Jan 2019 | San Francisco Bay Area Manager of the Video-on-Demand (VOD) Infrastructure team. In this role, I successfully grew the team from 4 engineers to 8 engineers, and was very proud of promoting 3 of them. The majority of the time was spent reducing our operational load and technical risks. Responsibilities include: * Interviewing and hiring for a variety of roles, including software engineers and product managers. * Mentoring and fostering career growth for members of my team. * Defining the annual project roadmap for the team. * Working closely with a variety of stakeholders to ensure team goals are aligned (including two directors within video, product managers, as well as sibling teams that used our services). * Defining technical strategy and providing feedback on architecture documents. * Communication and transparency on week-to-week operations and progress. * Attending organization-wide operational review meetings. * Defining KPIs and tenets for our team. * Business analysis of the cost model for the entire VOD system. Projects include: * Optimized the VOD CDN configuration to reduce monthly cost by $1MM. * Migrated a multitude of VOD post-processing jobs to a robust and reliable system using AWS best practices. * Designed and planned a replacement for the primary VOD backend service. * Variety of Twitch product features and optimizations. ### Software Development Engineer II @ Twitch Jan 2016 – Jan 2017 | San Francisco, California Senior contributor on the Video-on-Demand (VOD) at Twitch. The team was responsible for all backend VOD infrastructure and storage, VOD CDNs, Quality-of-Service for VOD playback, monitoring, video post-processors, and various product-related features. My role included bringing many of the best practices from AWS such as migrating to use more AWS services, continuous deployment, improved monitoring, and decision-making through the writing. Projects include: * Designed and delivered a Video Upload feature under tight timelines for TwitchCon. I was responsible for designing the API as well as implementing the backend infrastructure. The project was a success and shortly after launch scaled to ingest and transcode thousands of hours of video every hour. ### Software Development Engineer II @ Amazon Web Services Jan 2013 – Jan 2016 | Cape Town Technical lead for the newly formed EC2 Recovery Systems team. The team was tasked with solving the ever-growing problem of entropy within the EC2 fleet as it experienced thousands of host failures attributed to many different causes every day. Projects included: * Developed a new centralized system that automated various EC2 host failure scenarios such as rack power failures, rack network failures, hard drive failures, operating system bugs, EC2 lifecycle bugs, etc. The project intended to operate as a framework that could be plugged into by other EC2 teams to solve their own remediation issues. Versions 1 and 2 were developed in Java and AWS Simple Workflow. A third iteration of the framework used Ruby instead which allowed for a much more pluggable and de-centralized system that could be leveraged by all teams, as well as faster mean-time-to-resolution (MTTR). * Built a data ingestion pipeline that consumed various data sources into a single RDS SQL database to facilitate decision-making in the automated remediation system. Some data sources included EC2 software inventory, EC2 hardware inventory, EC2 instance health status, etc. The project started off using JRuby but we re-wrote the parsing and ingestion logic in Golang to be able to keep up with the large volumes of data. * Built a continuous deployment system for the recovery framework that leveraged best practices in order to deploy the software to the many combinations of AWS data centers and regions. The system was built on internal Amazon tools (Brazil and Pipelines), CloudFormation, as well as a variety of integration tests written in Ruby to ensure safe deployments. * Developed a highly sensitive automation system to reboot EC2 instances under tight timelines due to security vulnerabilities. ### Software Development Engineer I @ Amazon Web Services Jan 2010 – Jan 2013 | Cape Town Part of the backend EC2 team responsible for building, deploying to, and operating all software on the EC2 fleet. The complexity involved owning and operating many on-host software APIs and RPM packages that had to be continuously deployed to a fleet of millions of machines spanning many AWS data centers. Some projects include: * Expanding the on-host Ruby APIs to support new EC2 instance types using Xen (virtualization software) * Modifying the on-host Perl monitoring software to support multi-tenant configuration in order to publish metrics along team / project boundaries. * Many, many software deployments of a variety of RPM packages (third-party as well as software we owned) to millions of hosts across all the AWS regions. This included bug-fixing the deployment software written in Bash and Ruby, as well as remediating hosts that experienced a multitude of deployment failures. * Testing and validating all on-host software to operate with Amazon Linux instead of RHEL as we migrated the fleet to a new OS. * Many hours spent debugging all kinds of EC2 instance failures such as network attachments and blocked IO syscalls, broken hard drives, etc. * Bug-fixed the EC2 AMI launch caching algorithm that resulted in extremely slow launch times for large numbers of instances. * Primarily worked with EC2, EBS, SQS and S3. ## Education ### Master of Science (M.Sc.) in Applied Mathematics University of Johannesburg ### B.Sc. (Hons) in Applied Mathematics University of Johannesburg ### Bachelor of Science (B.Sc.) in Mathematical Sciences University of Johannesburg ## Contact & Social - LinkedIn: https://linkedin.com/in/jleroux0 --- Source: https://flows.cv/jaco JSON Resume: https://flows.cv/jaco/resume.json Last updated: 2026-03-31