Zircon

• Andrés Zunino, CEO: Civil Engineer with a Business Administration background. He partnered with Alejandro Narancio to launch ZirconTech in 2016 and leads global expansion.
• Martín Machín, Managing Partner: Mentor, investor, and advisor to high-growth technology companies; Managing Partner at Pyxis and former founder of Tata Consultancy Services' Global Delivery Center in Uruguay.
• Williams Olivera, CTO: Senior Software Architect leading ZirconTech's AWS-first strategy and flagship projects; founder of the blockchain practice.
• Marcelo Acosta, COO: Leads sales and operations, integrating AWS best practices and cost optimization.
• Alejandro Narancio, Board Member: Blockchain and decentralized-infrastructure leader who advises on blockchain strategy and security.

Offices and contact

• New York, US: 1325 Avenue of the Americas, New York, NY 10019
• Montevideo, UY: Bulevar España 2253, Montevideo CP 11600
• Email: contact@zircon.tech
• Phone: +1 (347) 354-2899

ZirconTech says it responds to project inquiries within one business day, or within 24 hours on the contact page. A 30-minute call can be requested for a high-level action plan and next steps.

Services

• AI Systems and Products: Agents, copilots, conversational interfaces, recommendation engines, LLM agents, RAG, and evaluation.
• AWS Cloud and Infrastructure: Landing zones, migrations, and cost-optimized architectures for AI workloads.
• Data and Machine Learning: Data platforms, predictive models, lakehouses, MLOps, and forecasting.
• Web3: Production-grade systems for fintech, tokenization, and digital platforms, including smart contracts.
• Human + Agent Teams: Senior engineers paired with purpose-built AI agents.
• Consulting and Strategy: AI readiness assessments, opportunity identification, and technology roadmaps.

Working approach

ZirconTech uses small senior teams, short feedback loops, and production-focused engineering. Clients work directly with the engineers making technical decisions. The company aims to produce a working slice in weeks, designs in observability, security, and cost controls, and delivers with measurable milestones. Everything built belongs to the client.

ZirconTech designs, migrates, and operates critical workloads on AWS, with an emphasis on faster releases, lower total cost, and stronger security. ZirconTech is an AWS Advanced Tier Services Partner.

Capabilities

• Landing Zone and Governance: Multi-account foundations with identity, logging, and cost tagging; landing zones can be delivered in three weeks.
• Migration and Modernization: Rehost, re-platform, or refactor workloads into containers, serverless, or managed databases.
• DevOps Automation: Infrastructure as code, CI/CD pipelines, and automated rollbacks.
• Data and AI Solutions: S3 data lakes, Glue ETL, SageMaker pipelines, and generative-AI agents.
• Cost and FinOps Reviews: Rightsizing, Savings Plans, and architectural adjustments; typical savings are stated as 15–30%.
• Security and Compliance: GuardDuty, IAM Access Analyzer, service control policies, and quarterly incident-response runbooks.

Cloud challenges addressed

ZirconTech addresses visibility gaps across accounts with centralized organization-wide observability, unexpected cost spikes with budget alerts and FinOps governance policies, cloud-team skill shortages with enablement workshops and hands-on knowledge transfer, and security-compliance complexity with pre-built SCPs, GuardDuty, and quarterly incident runbooks.

The website lists 25+ AWS certifications, three AWS competencies, a 99.9% uptime SLA achieved, and 15–30% typical cost reduction.

Delivery phases

Each package follows five phases:

1. Diagnosis: Process and data assessment, often funded by the AWS AI Assessment program.
2. Design: AWS architecture, integration plan, and success metrics.
3. Build: Implementation and integration with existing systems.
4. Go-live: Production testing and team training.
5. Operate: 30 days of post-launch support, followed by monthly operation.

The packages are based on problems documented across assessed companies and use fixed scope and fixed timelines.

• WhatsApp agent: Answers frequent questions automatically around the clock on WhatsApp.
• Voice agent: Handles inbound calls and routes escalations without hold time.
• Knowledge base: Uses internal documentation to provide accurate and consistent answers.
• Supervision panel: Provides real-time visibility into volume, response times, and cases requiring a human.

The solution is built on Amazon Lex, Amazon Connect, Amazon Bedrock, and Amazon S3. The delivery includes process and data assessment, AWS architecture design, implementation and integration, production testing, team training, and 30 days of post-launch support. After go-live, monthly operation is available.

An assessment may be available at no cost through the AWS AI Assessment program. Clients can then choose a pilot or full rollout.