EviSmart

Data Architecture Engineer

Most data roles give you one system to maintain. This one gives you the entire intelligence layer — every source, every schema, every signal — for a platform running live across 28 countries.

 

📍 BGC, Taguig, Philippines •  On-site  •  Full-Time

 

WHAT EVISMART IS BUILDING

EviSmart is the leading dental Autopilot system operating across 28 countries in a $40B healthcare industry — and the Manila team builds and ships it. Not an MVP. Not a pilot. Live, in production, with thousands of dental labs and dentists depending on the platform every day to run their business.

We have multiple in-house AI teams building the next generation of dental design automation — and Manila is where it gets operationalized. We are one of the few genuinely AI-first companies in the Philippines. The models are being built here. The product is scaling here. The opportunity is here.

 

WHY THIS ROLE, WHY NOW

The AI layer for the dental industry has not been built yet. The product is proven, but the roadmap is wide open. The Business Brain — EviSmart’s operational intelligence layer — requires a clean, governed, unified data foundation to surface insights, flag anomalies, and drive automated decisions. That foundation does not fully exist yet. This role builds it.

The person who joins now as Data Architecture Engineer doesn’t inherit a finished system. Every pipeline, every schema standard, every governance gate — they define what it looks like. The ML-Ops training pipeline, the Business Brain, and the KPI automation layer that the engineering team runs on are all downstream of this role. The data platform that gets built in the next 12 months is the infrastructure every AI and intelligence feature depends on. That architecture is yours to own.

Early joiners grow with the company. That is not a slogan — it is how every senior role here got built.

 

WHY EVISMART

• Leading dental Autopilot system in a $40B global healthcare industry — operating across 28 countries, live in production
• Multiple in-house AI teams building proprietary models — not just consuming APIs. Claude, Cursor, Luvable, and LLM-powered workflows are part of the daily stack
• Manila is not a back office. It is the delivery engine — operations, AI, product, and engineering all run from here. Your work ships globally
• We promote based on output, not tenure. The people who move fastest here are the ones who own their work completely
• 300+ people and growing — the career path is real because the company is growing into it

 

The data platform is one of the highest-leverage positions on the engineering team — and right now, it is being built from the ground up. We need someone who thinks in schemas before they think in code. Someone who loses sleep over an inconsistently named column not because they were told to care, but because they know that inconsistency compounds. If you are the kind of engineer who builds monitoring that catches the problem before even asked— that is who we want in this seat.

 

WHAT YOU’LL OWN

• Data Ingestion & Pipeline Reliability — own end-to-end ingestion from every source system (Customer-facing portals, Hive, LMS, HR Hub, HubSpot, QuickBooks, Jira, blob storage, and user interaction events) to the data warehouse. All pipelines run on schedule with >99% reliability. A missed or failed ingestion from any source is your problem to diagnose and fix before a downstream team reports it.
• Unified Data Warehouse — own the single repository where all organizational data lands, normalized and structured to a governed schema. Every data consumer — CS, RevOps, ML-Ops, the Business Brain — reads from the same warehouse. All schema changes go through you.
• Golden Record & Governance — own the Golden Record Checklist that every FRD must meet before engineering builds any feature that touches data. Your formal Data Architecture sign-off is a required gate — no build proceeds without it. Every entity (lab, case, file, user, interaction) has a canonical ID and a defined schema.
• ML-Ops & Business Brain Data Foundation — own the data layer that feeds AI and machine learning model training and powers the Business Brain’s operational intelligence output. Consistent schema, complete historical records, clean entity resolution. If an ML model produces unreliable results, the data layer is the first place to look — and you own it.
• Data Quality Monitoring — own the monitoring and alerting layer that catches quality issues before any downstream team discovers them. A quality issue reported by CS, RevOps, or ML-Ops is a monitoring failure, not a discovery mechanism.
• Data Platform Roadmap — you are the first to know when a pipeline is degrading, a schema decision is creating downstream fragmentation, or a new data source is being built around governance instead of through it. You act before it becomes a data incident or an engineering rework.

 

DATA SOURCES YOU OWN

Source	What You Capture & Why
Evident Design Portal	Case activity, lab interactions, file uploads, design workflow events, portal performance metrics.
Hive	Autopilot events, case booking timestamps, order flow, lab-level activity patterns.
LMS	Lab daily volume, revenue, products, rework and key lab information.
HR Hub	Workforce data, headcount, role assignments, shift patterns, attendance.
HubSpot	Every customer interaction — emails, deals, support tickets, churn events, MRR changes, CS classification records.
QuickBooks	Revenue, invoicing, payment status, MRR by lab, financial events.
Jira	Engineering work items, sprint data, bug tickets, deployment tags, cycle time, incident records.
Blob Storage / File Repositories	File IDs, metadata, upload timestamps, lab-to-file relationships, storage footprints.
User Interaction Events	Every action of every user across every system — logins, feature usage, session lengths, error encounters, API calls.

 

WHAT YOU’LL GET

• A seat at the AI table — work directly alongside in-house AI model teams using Claude, Cursor, Luvable, and LLM-powered workflows. The ML-Ops pipeline and the Business Brain are both downstream of your data platform. You are not support to those teams — you are their foundation
• Global reach, Manila-built — EviSmart operates across 28 countries. The data you capture and structure feeds intelligence that dental businesses in North America, Europe, and beyond depend on every day — from your desk in Manila
• Full data platform ownership — your pipelines, your schema standards, your governance gates, your roadmap. The single source of truth for a global platform is the system you define and defend
• A concrete career path — Data Architecture Engineer → Senior Data Architect → Head of Data Engineering. As the platform matures and the AI layer grows, the scope of this role grows with it
• Above-market compensation for data engineers who have personally owned pipeline reliability, warehouse architecture, and governance at a production scale. Range disclosed at first screen.
• HMO coverage, 13th month pay, and all government-mandated benefits

 

HOW WE WORK

We share work before it’s ready. We debate decisions loudly and execute them quietly. We default to fixing problems instead of escalating them. Manila is the delivery engine for a global platform — the standard is high and the pace is real.

A pipeline failure that a downstream team discovers before your monitoring does is not an incident — it is a system design problem. A schema change that bypasses your sign-off is not a shortcut — it is a governance failure. An FRD that touches the warehouse without your review is a build risk that someone else will pay for later.

If you’ve spent time waiting for someone else to define the data model — you’ll notice the difference on your first week.

Why EviSmart and not a bigger company? Because at a bigger company, your work goes into a queue. Your output is handled by a template. Here, what you build ships to 28 countries. That’s the kind of ownership most roles never give you — and we think the best people want it.

 

A NOTE ON WHAT WE’RE NOT

The most common question strong candidates ask about this role is: “Is this actually a data ownership role, or is it a pipeline maintenance function where someone else makes the architecture decisions?”

It is a fair question. The answer is that this role owns the architecture. The schema standards, the governance model, the Golden Record Checklist — you define them. The sign-off gate that prevents data-breaking features from going to engineering is yours to hold. The roadmap that matures this platform from its current state into the foundation for AI training and the Business Brain is your roadmap to write. The Head of Engineering Operations tracks milestones. The CTO reviews for strategic alignment. The architecture decisions are yours.

This is not a cost-center role and Manila is not a support function. Our Manila team leads engineering and operations — the product ships from here. If you’ve seen “Philippines office” mean something smaller at other companies, that is not what this is.

On AI: EviSmart’s AI is not a marketing claim. We have in-house model research and development teams, and the tools we use daily — Claude, Cursor, Luvable, LLM-powered workflows — are part of the actual stack, not a slide in a deck. The next 12 months will see the Business Brain’s operational intelligence layer and ML-Ops AI training pipeline built on top of the data platform this role owns.

 

WHAT WE NEED

• 3+ years building and maintaining production data pipelines end-to-end — not supporting someone else’s pipelines, but owning the reliability, schema, and quality of the data that lands in the warehouse
• Hands-on experience with ETL/ELT processes, incremental load patterns, and change-data-capture — you know when a pipeline is unhealthy before downstream teams notice, and you have the monitoring to prove it
• A track record of designing and owning warehouse schema — you understand the difference between a schema that works today and one that scales, and your schema changes are deliberate and documented
• Has run a data governance process before — naming conventions, onboarding checklists, sign-off gates — and treats governance as a quality mechanism, not bureaucratic overhead
• Communication that is clear, fast, and written — you can explain a complex schema decision to a RevOps analyst without technical jargon, and you can write a governance standard that other engineers will actually follow
• Familiarity with Fivetran or a comparable integration platform, experience designing data foundations for AI or ML use cases, and exposure to multi-source data environments (CRM, ERP, project management, and event streams together) are strong advantages

 

Apply at https://job-boards.greenhouse.io/evismart

EviSmart  •  Philippines  •  evismart.com