I'm Dan, a Principal Product Engineer based in Bucharest. I work across product design, architecture and implementation, and stay involved when the software reaches customers.
At New Yolk, I'm the co-founder and CTO and the sole engineer across web, mobile, integrations and operations. Artdevium is my independent company and the home for my own product work.
Based in Romania · Working internationally · Open to senior product engineering roles and direct collaboration, with scope and timing agreed around current commitments.
I turn business requirements into product decisions and deployed software. The responsibility includes customer journeys, the data behind them, integrations, analytics, releases and recovery.
Customer applications and staff operations meet in the same product model. A purchase needs to make sense on a phone, in an approval queue, in the ledger and in the report that follows.
The case study follows the architecture, money model, staff workflows, integrations and release decisions. It includes the constraints and corrections behind them, with smaller interactive examples where they help explain the behavior.
Read the complete case study →Customer web and mobile applications, staff workflows and a shared model of accounts, products and transactions.
Explore this work →Product-specific rates, conserved funding, delayed approvals and reversals tied to the original transaction.
Explore this work →Lead intake and distribution, ownership checks, approval workflows and reports with meaningful attribution.
Explore this work →Transactional outbox, bounded delivery retries and a deliberate limit on workflow orchestration.
Explore this work →Native sessions, customer languages, release-aware updates and verification of real failure paths.
Explore this work →I work through the customer journey and the data model together. That helps surface decisions that are easy to miss in a screen design: what the customer is actually confirming, what can change underneath it, and how the result will be explained.
My main stack is TypeScript, React, Next.js, Node.js, PostgreSQL and Expo. I also have Python familiarity and a background in JavaScript automation and ServiceNow.
I use Sentry and PostHog to understand errors and product behavior, alongside Vercel and Supabase operational tooling. My earlier work included SolarWinds and incident management.
AI-assisted development is part of my workflow. I still own the model of the problem, the review and the verification of what ships.
Public contributions make a different part of the work visible: focused changes, tests and review in someone else's codebase.
I've been coding and automating work since 2015. Full-time product software engineering became my primary work in 2022, building on years in support, incidents and platform delivery.
At Humans.ai, I built an AI media-to-NFT minting platform and an EVM-to-EVM bridge, prototyped a mobile crypto app, and shipped smart contracts and applications across Ethereum, Hedera, and Evmos.
The work included internal security reviews and coordinating external audits, alongside the applications and integrations themselves.
From L1 support onward, I built queue-alert scripts, triage tools, knowledge/search flows, ServiceNow workflows, dashboards, and automation while taking on incident, client, and service-delivery responsibility.
At GE I worked in the internal ServiceNow team. At DEVHD I managed ServiceNow delivery for Veoneer, Renault and Rolex while remaining hands-on with development and releases.
I built a private engineering system for work that crosses repositories and sessions. It keeps the request, decisions, current owner and evidence together so the next session can resume the actual work.
The hard part is deciding how much process a change needs. A small local edit should stay easy. A production change needs an explicit scope, a way to verify it and a recovery path.
Each objective has one typed, versioned record for lifecycle, ownership, authority, acceptance, budgets, decisions, and evidence. Tasks can change without creating a new objective or erasing history.
Reversible local work stays on a direct path. Cross-owner, production, external, destructive, privacy, security, financial, release, or migration work enters an assured path with explicit scope, verification, and rollback.
Every completed step must end in a verified state change, an explicit route or wait, or a terminal outcome. Status messages and generated views cannot become proof or authority.
Optimistic versions, owned-input revisions, idempotency, append-only transitions, stop conditions, and rollback plans make collisions and retries visible before they compound.
Sanitized architectural evidence only. Source, internal records, identities, credentials, customer data, provider details, and exploitable control information remain undisclosed.
This is internal operating software, not a claim of commercial adoption or measured productivity gains.
The tooling supports the product work. It is one part of my engineering practice, alongside the applications and services customers use.
If you need a senior engineer who can work through product decisions and take responsibility for delivery, send me the context. I'm open to senior roles and direct product engineering engagements.
New Yolk is an active commitment. We can discuss scope, working arrangements and timing rather than assume immediate availability.
danvdoca@gmail.com