Projects
Selected projects
AI architecture and system design for applications, platforms, and critical operational software.
The common thread across all of them: trust as a property of the system, earned in how architecture is decided, how context is handled, and how AI reaches the user.
TÜV NORD Argentina · Sistema Cóndor
2023–2025 · Enterprise platform
TÜV NORD operated with manual processes scattered across multiple tools. Generating a commercial proposal took days of manual work, with no centralized visibility into certifications or audits.
The problem
Users with very different expertise levels interacting with intelligent automation. The system had to decide when to act alone and when to request human validation — without frustrating experts or abandoning less technical users.
What I designed
Full operational platform with AI integrated from the architecture. Agents generate pre-proposals as editable drafts — the user always stays in control. Flows show what the system did and why. Validation rules are configured by role. From 3 days to 5 minutes in incoming query evaluation.
Key results
80% less administrative load
100% of operations centralized
2 years in production
Mi Caja · Government Platform
2019–2023 · Platform at scale
Managing administrative procedures for retired personnel depended on in-person visits and manual coordination between agencies. No one had real visibility into the status of each case.
The problem
Processes dependent on in-person presence, manual coordination between agencies, and zero visibility into the real status of each case.
What I designed
Digital platform that centralizes procedures and coordinates approval flows between agencies. Web portal, mobile app, and complete inter-institutional workflow. Security and traceability for a critical system with more than 100k active users.
Key results
70% reduction in processing times
100k+ active users
Full inter-institutional traceability
B2B Industrial Support Tickets
2025 · In development · AI agents · Context engineering
The bottleneck wasn't ticket volume. It was what happened inside each one: incomplete descriptions, back-and-forth rounds, technicians starting from scratch each time.
The problem
Different roles needed different interactions with AI, and the confidence threshold to act on suggestions was different for each one.
What I designed
Two agents with distinct identities — one for the operator under pressure, another for the expert technician — connected by a structured context object. Each suggestion includes a confidence level and traceable source. The system knows when to escalate and transfers complete context to the human, with no loss.
Projected results
50% projected reduction in resolution time
75–88% of context pre-filled automatically
Eunoia Care · AI Assistant for Caregivers
2025–present · Healthtech · In development · Multi-agent system
Families caring for a chronic patient live with fragmented medical information: tests, hospitalizations, instructions from different specialists. Forgetting and fragmentation are the main source of anxiety and errors.
The problem
Designing an experience where a caregiver without clinical training could capture information without friction and recover it when needed. Trust was the central design constraint.
What I designed
Multi-agent system with distinct specialties — lab interpretation, longitudinal medical history, natural language conversation — each with its own context, tone, and autonomy limits. The caregiver captures in any format and the system responds with data or honestly indicates when it doesn't have the information.
Design results
Significant reduction in cognitive load
Longitudinal medical history built through use
Medical information available in seconds
Have an AI project in mind, or a system that needs more rigor in its AI layer?
Tell me what you're building and I'll respond within 24 hours.