Alexis Mas

Alexis Mas

Infrastructure & platform engineering

CV

Senior Infrastructure Engineer · Valencia, Spain · Remote (CET · full EMEA overlap · afternoon overlap with the US East Coast)

[email protected] · LinkedIn · GitHub

Infrastructure and platform engineer with 15+ years of experience running production Kubernetes (EKS) and AWS for high-traffic products. I own reliability end-to-end (Terraform and Pulumi, GitOps delivery, observability, on-call and incident response, cloud cost), and I build internal developer tooling as a product: requirements gathered from the engineers who use it, then documented and driven to adoption through workshops, written guides, screencasts and conference talks.

Six years fully remote in small, autonomous, async teams, all of it with US companies and colleagues on both coasts, most recently as the sole platform engineer on a multi-terabyte data platform. Currently focused on AI-assisted infrastructure work: agent-ready development environments, tight feedback loops, and automating recurring operational toil.

Experience

Coinwatch — Senior Software Engineer, Infrastructure

Nov 2025 – Present · Remote (Spain) · Contract · US company

AWS (ECS, EC2, Aurora/RDS, S3, ECR, SSM, VPC) · Pulumi (TypeScript) · Datadog · Ruby on Rails · Sidekiq · Docker · CodePipeline/CodeBuild · Cloudflare

Sole platform engineer for a crypto market-data platform: multi-terabyte Aurora, continuous ingestion from exchanges and on-chain sources.

  • Piloted autonomous AI-assisted development on the platform: built a devcontainer-based environment to let agents run with reduced blast radius, supported by tight feedback loops and thorough agent-focused documentation. Now used for day-to-day delivery.
  • Built Datadog observability from scratch (APM, logs, metrics, monitors), replacing a fragmented setup; extended coverage to AWS Nitro Enclave workloads via Vector.
  • Responded to production incidents: diagnosed and resolved recurring OOMs and a site-wide outage, tracing both to specific upstream data conditions.
  • Introduced Pulumi IaC with AWS OIDC; imported existing hand-managed resources into managed stacks.
  • Identified SQL patterns driving ~33% of database cost; cut worst-case query time from 4h to 20s with composite indexes and aggregate rewrites.
  • Cut AWS spend by 10%: EC2 right-sizing, CloudWatch rationalisation, ECR lifecycle policies, Aurora migrated to Graviton, and recurring load spikes flattened by redistributing cron schedules.
  • Reduced web tail latency 75% (20s → 5s) and halved Sidekiq infrastructure through instance-type and queue tuning.

Wellfound — Senior Software Engineer, Cloud Infrastructure

Jan 2020 – Oct 2025 · Remote (Germany / Spain) · US company

Kubernetes (EKS) · Terraform · AWS · ArgoCD · Helm · Datadog · Cloudflare · Buildkite · Docker · Ruby on Rails · GraphQL · Elasticsearch · Redis · MySQL · Postgres · Python / Dagster

One of two infrastructure engineers supporting 20+ product engineers across US and European time zones, owning reliability, scalability and delivery pipelines for a platform serving 3M+ monthly visitors.

  • Operated and upgraded 3 production EKS clusters, plus Ruby, Rails and Elasticsearch, without downtime; managed cluster dependencies and release lifecycle with Helm.
  • Ran GitOps delivery with ArgoCD; set up canary deployments for the main application.
  • Built near-instant rollback into the data team’s Dagster pipeline (Python): clone the ~1TB Elasticsearch index before each multi-hour run, repoint the alias on failure. Recovery went from a 4h+ snapshot restore to seconds, letting the data team ship changes without risking a corrupted production index.
  • Shared a week-long on-call rotation with the other infrastructure engineer; authored runbooks and postmortems. Drove incident frequency from monthly to a few per year through caching strategies, DB read/write splitting and resource tuning.
  • Managed all infrastructure as code in Terraform: split a monolithic workspace into smaller, well-scoped workspaces with clear ownership and blast-radius boundaries; cut infrastructure costs by ~10%.
  • Migrated observability from ELK to Datadog, designing dashboards and alerting; ran internal workshops to upskill the engineering team.
  • Optimised CI/CD pipelines, cutting build times by 80% and 50% on the two most business-critical applications.
  • Reduced key endpoint latencies from seconds to sub-100ms (GraphQL overfetching across datasources, N+1 queries, missing indexes).
  • Drove SOC 2 compliance, getting the company attested in a couple of months vs. the typical 6–12 month timeline.

New Work SE — Senior Software Engineer, API Platform

Dec 2016 – Dec 2019 · Germany

Scala · GraphQL · Ruby on Rails · Kubernetes · Docker · Prometheus · RabbitMQ

First engineer on the platform team that gave product teams the infrastructure to expose their APIs. The team grew from 2 to ~8 engineers during my time. Owned the existing REST gateway while designing and building its GraphQL successor.

  • Designed and built a new GraphQL API gateway in Scala, learning both Scala and GraphQL on the job and prototyping in Elixir. It became core infrastructure adopted by multiple teams (100+ engineers).
  • Ran the gateway as an internal product: features shipped because product teams needed them, backed by extensive documentation and onsite workshops delivered across offices in Germany, Spain and Portugal.
  • Presented the platform work publicly at a meetup in Germany and a conference in Portugal.
  • Created internal developer content (written guides plus screencasts and feature demos on the internal blog) to scale adoption past what in-person support could reach.
  • Onboarded product teams progressively rather than all at once, keeping platform-team bandwidth realistic and the tooling solid at every step.
  • Maintained the legacy Rails REST gateway serving 10k+ req/s throughout the transition so teams could migrate at their own pace, and migrated it to Kubernetes with zero downtime.

WyeWorks — Software Engineer

May 2015 – Nov 2016 · Uruguay · US-based clients

  • Built Rails-based web applications for US-based startups, focused on clean architecture and performance.

Earlier

2008–2015: Software Engineer at Scanntech, Holla@Me, Crambo and Universidad de Alcalá (mobile & web applications).

Technical skills

  • Cloud & IaC: AWS (EKS, EC2, S3, ECR, IAM, ALB, RDS/Aurora, SSM, VPC) · Terraform · Pulumi · Cloudflare
  • Containers & Delivery: Kubernetes / EKS · Helm · ArgoCD · GitOps · canary & progressive delivery · Docker
  • CI/CD: Buildkite · GitHub Actions · CodePipeline/CodeBuild
  • Observability & Reliability: Datadog · Prometheus · Rollbar · on-call & incident response · runbooks & postmortems · SOC 2 (Vanta)
  • Data: Postgres / Aurora · MySQL · Elasticsearch · Redis
  • Edge & Security: Cloudflare · DataDome · AWS OIDC · Secrets Manager / 1Password / dotenvx
  • Languages: Ruby · Python · TypeScript · Scala · Java · Bash

Education

Universidad de Alcalá — Bachelor of Engineering, Computer Science · 2005–2012 · Spain