CDI

Senior ML Platform Engineer

QualityAI 1424 Buenos Aires

Contrat
CDI · Temps plein
Lieu
1424 Buenos Aires
Diplôme souhaité
Indifférent
Expérience
2 à 4 ans
Postuler à cette offre Réf. 8145425

Description du poste

QualityAI is an AI-first quality engineering company helping enterprises deploy and scale complex systems with greater confidence. Operating across data, models, platforms, infrastructure, and operational environments, the company provides assurance and engineering expertise that helps organizations ensure systems perform reliably in real-world conditions. Formerly Qualitest, QualityAI supports global enterprises across regulated and technology-driven industries, combining deep engineering heritage with AI-enabled delivery, operational assurance, and lifecycle expertise to help clients achieve certainty at go-live. Are you interested in working with the World’s leading AI-powered Quality Engineering Company? Ready to advance your career, team up with global thought leaders across industries and make a difference every day? Join us at QualityAI! We are looking for a Senior ML Platform Engineer to join our growing team in Argentina/Mexico (100% remote)!    Location: Argentina/Mexico Remote Time zone: MST ideal but we can also do CST or EST-   Position Overview We are seeking a Senior ML Platform Engineer to build and operate the high-throughput, agentic pipelines that process millions of historical documents. You will turn ML prototypes into scalable, cost-optimized, and resilient production systems on AWS, and keep them healthy once they're live. About a third of this role is keeping production healthy: AMI updates, security patching, incident investigation, and cost cleanup. If you enjoy owning systems in production, you'll like it here. This is an Engineering role, not a Data Science role. You won't train models or design prompts. Our Data Science team hands you validated models and LLM prompts; your job is to make them run reliably and cheaply at scale. You'll join our Applied AI Engineering team, working alongside Applied AI Scientists. They build and validate the models; you build and run the systems that put those models into production. Profil recherché: Key Responsibilities * Production Pipeline Engineering: Build and maintain scalable pipelines (EC2, Lambda, S3, SQS) that execute agentic workflows for entity extraction. * Model Serving: Deploy and manage open-source LLMs on EC2 and integrate third-party LLM APIs, focusing on throughput and reliability. * Inference Tuning: Tune vLLM serving parameters for throughput and cost, including serving LoRA adapters. * GPU Capacity & Cost: Manage GPU capacity across AWS regions using Savings Plans, spot, and on-demand instances to keep unit cost down while securing enough GPUs. * Production Operations: Own ongoing maintenance of production stacks: AMI updates, security patching, vulnerability scanning (Snyk), and stack retirement. * Observability & Reliability: Implement monitoring, alerting, and automated error-handling/retry logic; troubleshoot and resolve production pipeline issues. * Engineering Partnership: Act as an engineering contact for the Data Science team, operationalizing their models and prompt logic and advising on hosting, cost, and operations. * Infrastructure as Code: Manage the AWS infrastructure supporting the ML lifecycle. * Documentation: Write clear technical documentation and architecture diagrams.   Required Technical Skills * Infrastructure as Code: Proficiency with Terraform or similar tools for modular, version-controlled AWS infrastructure. * Python (Systems Level): Advanced proficiency building robust, multi-threaded, or asynchronous applications. * CI/CD: Harness or similar deployment pipelines. * Operational Excellence: Proven track record in production monitoring (New Relic, CloudWatch), logging, alerting, and troubleshooting distributed systems. * Security: Strong grasp of AWS IAM, security patching, and vulnerability remediation. * LLM Serving: Working knowledge of how LLMs are served and called; hands-on experience with vLLM or similar inference servers. AI-Assisted Engineering: Daily, hands-on use of AI coding tools (e.g., Claude Code, Kiro). You know how to make them effective (clear context, well-scoped tasks, reusable instructions) and when not to trust them: you verify output, catch confident mistakes, and own every change you ship.   Preferred Qualifications * MLOps Focus: Experience with CI/CD for ML, containerization (Docker), and automated testing of data pipelines. * High-Volume Processing: Experience with OCR or document digitization at scale. * Throughput Tuning: Experience managing rate limits and costs across multiple LLM providers. * Multi-Region: Experience with multi-region AWS deployments.   Must haves: * Hands-on AWS production experience (EC2/Auto Scaling, Lambda, S3, SQS), Terraform, and Python * Comfortable with operational work: about a third of the role is maintenance, patching, and investigating production issues   What You'll Work On * The Engine: Building the core "machinery" that digitizes and analyzes vast historical archives. * Scale: Moving from "proof of concept" to processing large volumes of documents concurrently. * Resiliency: Building systems that can gracefully handle API outages, model timeouts, and malformed data.   Impact You'll own the systems that turn validated models into production throughput. Every gain you make in reliability, speed, or unit cost multiplies across millions of documents, and determines how much of the world's historical archives we can afford to make searchable.

Profil recherché

  • Études / diplôme : Indifférent
  • Expérience : 2 à 4 ans
Ce poste vous intéresse ?

Envoyez votre candidature directement au recruteur.

Postuler