Disclaimer – Must read: Commitment & FocusThis role requires full-time dedication, with clear priority given to Darwoft projects during the established working hours. It is not compatible with other full-time professional engagements. Any additional professional activities must be disclosed in advance and must not interfere with the responsibilities or working hours of this role.
Darwoft is a software factory that develops custom software solutions and provides IT staff augmentation services for international clients, primarily in the United States and Latin America. We work with startups and high-growth companies to build high-impact digital products. Our culture is people-first, focused on technical quality, long-term relationships, and a collaborative mindset. We combine technical excellence with human proximity.
-
Location: Remote (Global)
-
Contract Type: Contractor
-
Industry / Project: AI Infrastructure & Platform Operations
-
Time Zone: Coordination with US / LATAM teams
-
English Level: Advanced (C1)
We are seeking a Senior Site Reliability Engineer with a deep focus on observability and AI platform operations. This role sits at the intersection of reliability engineering and emerging AI infrastructure. You will own the instrumentation, visibility, and operational health of AI-powered systems, including LLM API gateways, token usage pipelines, and model serving infrastructure.
You will act as the authority on runtime behavior across our AI stack, building the tooling and insights required to understand, measure, and optimize system performance, reliability, and cost.
-
Design and operate AI gateway infrastructure, including routing, rate limiting, and traffic shaping for LLM API traffic.
-
Build and maintain deep observability into AI workloads: token consumption, model latency, cost attribution, and error rates by model, team, and use case.
-
Define and track SLIs, SLOs, and error budgets for AI services and API-dependent workflows.
-
Instrument LLM-backed applications to surface prompt/completion telemetry, retry patterns, and quota burn rates.
-
Develop dashboards and alerting using Grafana, Loki, and Prometheus tailored to AI traffic patterns (beyond traditional infrastructure metrics).
-
Maintain and evolve observability pipelines capable of handling high-cardinality AI metadata.
-
Lead incident response for AI platform degradations, including model unavailability, gateway saturation, and upstream provider outages.
-
Automate operational workflows across AI infrastructure using Infrastructure as Code (IaC) and CI/CD practices.
-
Collaborate closely with ML/AI engineering teams to embed reliability and cost-visibility practices early in the development lifecycle.
Must-Have:
-
Strong experience with AWS cloud services, specifically those relevant to AI workloads (Bedrock, SageMaker, Lambda, API Gateway).
-
Hands-on expertise with Kubernetes in production environments.
-
Proven experience building and operating observability stacks (Prometheus, Grafana, Loki), with an emphasis on application- and API-layer metrics.
-
Solid understanding of API gateway patterns, including routing, throttling, authentication, and traffic observability.
-
Experience instrumenting and monitoring LLM or AI API usage (token budgets, cost tracking, latency profiling).
-
Proficiency in Python, Go, or Bash for automation and tooling.
-
Mastery of Infrastructure as Code (Terraform) and CI/CD pipelines.
-
Strong analytical mindset with the ability to extract signal from high-cardinality telemetry.
Nice-to-Have:
-
Experience operating or integrating AI gateway solutions (e.g., Kong AI Gateway, Portkey, LiteLLM).
-
Familiarity with OpenTelemetry and distributed tracing for AI/ML workloads.
-
Experience with FinOps practices for AI, including chargeback models and cost anomaly detection.
-
Knowledge of service mesh technologies and their role in AI traffic management.
-
Contractor agreement with payment in USD.
-
100% remote work in an international environment.
-
Access to Argentine public holidays.
-
Professional development in the cutting-edge field of AI Platform Engineering.
-
Referral program and access to learning platforms.
-
English classes to further enhance professional communication.
Explore this and other opportunities at:www.darwoft.com/careers
