Technology Leader / Systems Architect / Product & Engineering Leadership
Belgrade, Serbia
Email: lordzlis@gmail.com · Phone: +381 62 979 5315 · LinkedIn: https://www.linkedin.com/in/nazarovanton/
Languages: Russian native · English C1 / fluent · Serbian working proficiency
Portfolio: Portfolio Index
Technology leader and systems architect with 18+ years across startups, enterprise software, banking, public-sector systems and founder-led businesses.
I am usually most useful when the technology already exists but the work around it is messy: part of the process lives in Excel, part in email or chat, access rules have drifted, old architecture decisions are undocumented, and people no longer agree on who owns which step. I start by mapping what people actually do, where time or data gets lost, which responsibilities are real, and what can be changed without breaking daily operations.
Some of my roles have been direct line management; others combined architecture, product ownership and cross-organizational authority. My largest verified direct team was 5 IT staff and 2 installation engineers. Separately, at the Directorate of Cultural Centers I coordinated technology requirements and operating standards with 100+ IT specialists across 100+ institutions without line authority. I keep those two scales separate because they are different kinds of management experience.
I stay close enough to implementation to challenge architecture, platform and vendor choices, but my strongest work is usually one level above the tools: team design, operating model, economics, security, process ownership, adoption and deciding what should be built, bought, reused, stopped or simplified.
This is the evidence-rich master resume. It is intentionally longer than a vacancy-specific CV; targeted versions should select from it rather than trying to preserve every signal.
- Largest verified direct team: 5 IT staff + 2 installation engineers at ZIL, with differentiated ownership across service desk/documentation, workplace technology, infrastructure/DevOps, business systems and physical technology.
- Indirect leadership at scale: coordinated requirements, standards and delivery with 100+ institutional IT specialists across 100+ cultural institutions without direct reporting authority.
- Hiring and team building: hired, onboarded, trained and mentored employees and interns; developed people from both conventional and nontraditional backgrounds when their capability matched the work.
- Performance management: introduced explicit responsibilities, measurable service expectations and KPI-based management; made personnel changes when inherited arrangements did not meet operating requirements.
- Organization design: combined full-time, fractional and shared specialist capability rather than staffing every competency as a permanent full-time role.
- Delegation: typically diagnose and design the operating model, launch the work with the relevant specialist, document the process and then delegate steady-state execution while retaining accountability for economics, risk and priorities.
- Durability: the ZIL service/technology operating model continued working for approximately 18 months after my departure.
- Executive stakeholders: worked directly with founders, directors, department heads, bank leadership, public authorities, vendors and external technology organizations; represented ZIL in higher-level organizations including Moscow DIT.
- Cross-functional ownership: technology work has covered finance, accounting, legal, procurement, HR, service operations, events, customer/client processes and public-sector reporting rather than remaining inside an IT department.
- Helped a service business achieve approximately 10x revenue growth in about three months after redesigning customer intake, operations and product presentation; growth was organic, without paid advertising.
- Reduced routine client-service work from hours to minutes through structured systems, integrations and automation.
- As part of the ZIL management team, supported an anti-crisis turnaround from approximately RUB 30m in accumulated debt through internal optimization, automation and tighter technology economics.
- Avoided approximately RUB 22m in potential workstation replacement and licensing costs by choosing lifecycle extension and licensing strategy over blanket replacement.
- Rebuilt ZIL support from a largely unmanaged service into a measurable one: routine issues were often resolved in approximately 10–15 minutes, response and resolution SLAs were introduced, and monthly demand fell from roughly 300–600 requests to about 150 over approximately 6–12 months as backlog, workstation inconsistency and recurring user issues were reduced.
- Moved 100+ employees to secure remote work during COVID while maintaining operating continuity.
- Found 160+ accounts/mailboxes associated with former employees still active after departure and rebuilt onboarding/offboarding and identity-lifecycle controls so access removal became part of the HR/IT process.
- Converted license and certificate management from surprise expiry and business interruption into a planned renewal/procurement cycle with advance reminders.
- Replaced fragmented phone/email/Excel coordination for event support with one tracked workflow across at least four delivery functions, including shared scheduling, ownership and equipment reservation; approximately 300 events passed through it during practical use.
- Built AIS MosRazvitie into a sector-wide operational source of truth across 100+ cultural institutions and libraries. Its data was used for management planning and resource allocation; event volume recorded in the system influenced planning and funding.
- Built a banking document-management system used by approximately 60 back-office users; after rollout documents stopped disappearing inside the internal process, ownership and overdue status became traceable, and routine internal courier movement was no longer needed.
- Helped stand up the IT function and workplace environment for approximately 100 users in a newly created organization under an ASAP operating mandate.
- Consolidated field operations and client workflows while migrating approximately 70,000 historical documents into a structured backend.
- Supported and developed an enterprise IBM Domino document environment used by 3,000+ users at Beluga Group earlier in my career.
A recurring part of my work is telling people that the technically possible choice is not necessarily the sensible one. I look at supportability, adoption, internal capability, future change and exit cost, not only the feature list or license price.
- At ZIL, stopped further investment in Bitrix24 after approximately RUB 90k had already been spent when testing showed the platform was too complex for the actual user base; shifted process work toward Pyrus instead of defending a sunk-cost choice.
- Evaluated Odoo as a self-hosted/free option but rejected it when infrastructure demand, customization requirements and the need for specialized Python development made the real ownership cost unattractive.
- Advised clients away from Glide when customization ceilings constrained the product; moving to WeWeb enabled map-heavy interfaces, themes, better authentication flows and a proper desktop experience.
- Recommended Supabase instead of Airtable for a structured backend. That backend has remained in production for approximately three years and survived a complete frontend rewrite without replacement of the core data layer.
- Delivered a production system on Xano even though I had recommended another direction. The architecture and documentation were explicit enough that the client later replaced the entire Xano backend with an agentically developed alternative.
- Prefer pilots and progressive rollout before major commitment. I treat adoption and operational fit as investment criteria rather than assuming that a product with more features is automatically the better choice.
- Have rewritten failed externally developed systems when continuing to patch them would have cost more than replacing the underlying design.
Founder / CTO / Systems Architect / Project Manager
Nov 2022 - Present
I run an independent technology practice where projects are rarely just “build an app.” Usually there is an operating problem behind the request: a team losing time in chats and spreadsheets, a founder who cannot see what is happening in the business, a platform that has hit its ceiling, or an old implementation that needs to be replaced without stopping work. I cover architecture, delivery, cleanup, automation and the first adoption/support loop after launch.
- Work directly with founders and business owners, translating operating constraints into architecture, product scope, automation priorities, ownership boundaries and delivery trade-offs.
- Design and deliver CRM, ERP, internal portals, reporting, marketplace, assessment and field-operations products across PostgreSQL, Supabase, WeWeb, Xano, REST APIs, n8n, Make, Telegram integrations and AI APIs.
- Redesigned customer intake, booking, communication and internal workflows for a service business; the broader restructuring supported approximately 10x revenue growth in about three months without paid advertising.
- Reduced routine service handling from hours to minutes by replacing repeated manual handoffs with structured workflows, integrations and automation.
- Consolidated field operations, client workflows and reporting in one platform and migrated approximately 70,000 historical documents into a structured backend.
- Design systems around explicit workflow states, permissions, data relationships, notifications and reporting rather than burying business rules in front-end screens.
- Reduce vendor and key-person dependency by separating data, business logic and interfaces and documenting ownership and handover paths.
- Designed a Supabase backend that has remained in production for approximately three years and survived a complete frontend rewrite without replacement of the core backend.
- Delivered a production Xano-based system whose full backend was later replaceable from the architectural documentation, demonstrating handoff quality and portability even when the original platform was not my preferred choice.
- Moved client work away from Glide where customization limits prevented richer map, authentication, theme and desktop experiences.
- Build AI-assisted engineering workflows using repository instructions, bounded tasks, acceptance criteria, architectural guardrails, execution checkpoints, review loops, auditable SQL practices and database-to-git workflows.
- Built PromptlessPress as a structured AI generation system with reproducible prompt composition, execution snapshots, debugging flows and shared backend helpers rather than a thin prompt wrapper.
Instructor
Sep 2023 - Dec 2023
- Taught practical topics around Bubble development, Figma/product interface work, AI tools and international marketplace positioning.
- Turned applied product and low-code delivery experience into lessons, examples and student-facing explanations.
Implementation Curator / Product Manager
Aug 2022 - Nov 2022
Worked inside a commercial automation company implementing ERP/CRM systems and looked at the part that is easy to miss: what happens after a technically finished implementation reaches the client.
- Investigated why technically completed ERP/CRM implementations were failing to become working daily management tools.
- Connected failure to unclear ownership, client communication, post-launch responsibility and implementation method rather than assuming missing features were the root problem.
- Helped shift implementation success from technical acceptance toward actual adoption, managerial ownership and measurable business effect.
- Worked with founders and managers on implementation methodology, role boundaries and post-delivery operating responsibility.
Head of IT
Mar 2020 - Jun 2022
I joined the management team during an anti-crisis period. The immediate technology mandate was to reduce operating cost, close security and service gaps, and create enough trust in IT to use technology outside the IT department. The longer-term work was to help make ZIL a more technologically progressive and modern cultural venue.
- Managed 5 IT staff and 2 installation engineers plus vendors and contractors.
- Rebuilt the inherited support model around differentiated specialist ownership rather than one generic support role: service desk/documentation, on-site endpoint work, deeper remote infrastructure/DevOps expertise and physical-technology support.
- Used fractional senior DevOps capability where the organization needed deeper expertise but did not need or could not economically justify a full-time equivalent role.
- Rewrote responsibilities and operating rules, introduced measurable service expectations and performance management, and made personnel changes where inherited performance did not meet requirements.
- Hired and developed staff from both conventional and nontraditional backgrounds, using training and role design to match people to the work rather than filtering only by prior title.
- Designed backup coverage around critical functions without turning everyone into interchangeable generalists.
- Built a service/technology operating model that continued functioning for approximately 18 months after my departure.
- As part of the management team, supported turnaround from approximately RUB 30m accumulated debt through internal optimization, automation and tighter operating economics.
- Chose lifecycle extension and licensing optimization over blanket workstation replacement, avoiding approximately RUB 22m in potential replacement and licensing costs.
- Managed the full IT procurement lifecycle under 223-FZ: supplier/solution research, technical requirements, procurement coordination, acceptance, implementation through the team and final accountability for the result.
- Evaluated procurement by total organizational cost rather than unit price alone; did not split purchases when the extra tender effort and delay would cost more than the price difference being optimized.
- Negotiated vendor discounts and trial periods before larger commitment where possible.
- Stopped further investment in Bitrix24 after approximately RUB 90k sunk cost when user testing showed poor adoption fit, and shifted process work toward Pyrus.
- Replaced support through corridor conversations, calls and personal messages with a managed HelpDesk, documented request types, ownership, urgency/business-impact prioritization and explicit response/resolution SLAs.
- Routine issues were often resolved remotely in approximately 10–15 minutes; monthly request volume fell from roughly 300–600 to about 150 over approximately 6–12 months as backlog was cleared, endpoints were standardized and repeated user issues were reduced through training.
- Used service analytics to identify recurring demand and separate genuine system failures from training/process problems rather than treating ticket volume itself as success.
- Centralized working data and standardized endpoints so a failed workstation could be swapped and the employee could continue under the same identity and centrally stored data instead of depending on one physical PC.
- Moved 100+ employees to secure remote work during COVID without interrupting core operations.
- Recovered critical control after a key-person continuity failure, then centralized privileged access, documentation and ownership so core systems no longer depended on one administrator.
- Audited personal-data handling, information security, access, licensing, service operations and integration risks across the organization.
- Found more than 160 former-employee accounts/mailboxes still active and rebuilt identity lifecycle/offboarding so account removal became a controlled HR/IT process.
- Defined role-based access and endpoint-control models; the team standardized permissions, shared storage, domain configuration, public-computer privacy controls and monitoring.
- Moved license, key and certificate tracking into GLPI with advance reminders so renewal/procurement work started before expiry rather than after interruption.
- Rebuilt technology governance for event support, equipment custody and accountability.
- Restored and expanded CCTV/retention and access-control processes; recordings supported approximately three theft investigations and property recovery in some cases.
- Extended the service-management pattern beyond IT into additional internal service functions.
- Built one tracked event-request workflow covering at least IT, lighting, sound and installation/technical setup, replacing fragmented email/phone/Excel coordination with a shared calendar, equipment reservation, history and explicit ownership; approximately 300 events passed through the process.
- Worked with finance/accounting, legal, procurement and technical operations to move financing requests, contract/procurement initiation, procurement-document workflows and service requests into structured digital processes.
- Led structured CRM selection through requirements discovery, market research, candidate comparison, user pilots and operational-fit testing.
- Supported launch of paid event streaming during COVID as a new digital event capability.
- Contributed product and technology direction for the new website and broader modernization intended to reposition ZIL as a modern cultural venue.
- Represented ZIL in higher-level organizations including Moscow DIT and chose centrally provided/shared technology where reuse was economically and operationally preferable to duplicating capability in-house.
Senior Manager / IT Architect / Analyst
Jan 2017 - May 2023
Long-running cross-institutional role held concurrently with later institutional leadership appointments.
This was a hybrid role: part architect, part analyst, part internal product owner and part cross-institutional technology coordinator. The systems served Moscow's cultural sector, so technical work also meant reporting obligations, access rules, user adoption, support and coordinating people in institutions that did not report to me.
- Coordinated technical requirements, operating standards, implementation decisions and priorities with IT directors, heads of IT and specialists across 100+ institutions.
- Prepared sector-level technical requirements and represented institutional technology needs in meetings with public authorities.
- Participated in centralized procurement planning and quota allocation, consolidating institutional needs and tracking delivery.
- Hired, onboarded, adapted and mentored employees and interns and designed probation/onboarding processes.
- Owned product logic, architecture, data model, workflows, access rules, reporting, governance, implementation and adoption of AIS MosRazvitie.
- Turned a widely disliked legacy system into the operational source of truth across 100+ cultural institutions and libraries.
- Started by interviewing departments about what they actually did with the data. The old model did not match reality: one institution could occupy multiple buildings, one building could contain spaces assigned to different institutions, and events, staff and sections needed more flexible relationships than a rigid hierarchy allowed.
- Expanded the platform into an ERP-like operational and management information system covering organizational structure, spaces, staff, vacancies, salaries/finance where appropriate, clubs and sections, events, accessibility, equipment and infrastructure, city-program participation, KPI and planning data.
- Supported an event lifecycle of plan -> approval -> assignment of responsible staff -> delivery -> report/attendance.
- Implemented role-based visibility so leadership, finance, event teams and section managers saw and edited only the information appropriate to their work.
- Introduced field-level data freshness and completion logic so management could distinguish current from stale information and institutions could update complex records progressively instead of losing partial work.
- Made urgent equipment/event questions answerable from centralized data instead of starting a new manual collection from institutions each time.
- System data supported management planning and resource allocation; event volume influenced planning and funding.
- Replaced one major events report that took approximately 5 hours and was effectively limited to one run per day with server-side reporting completing in seconds to minutes.
- After redesign, logs showed approximately 200–300 report runs per working day across central and institutional users. I treat this as evidence of self-service adoption and throughput, not as a high-load claim.
- Institutions reused the same reporting layer with visibility restricted to their own organization, including exports for downstream work.
- Designed platform mechanisms that reduced change lead time: new fields and sections could often be added within roughly a day instead of requiring a long custom-development cycle.
- Phone support could tie up one specialist with one user for close to an hour while providing very little context. I moved support to Tawk.to, where the team could work with roughly 10–15 sessions in parallel and diagnose issues with much better user/context and screen visibility.
- Used recurring documentation, consultations and BigBlueButton/webinar training for hundreds of users so new product blocks, reporting rules and data-quality requirements were learned as part of rollout rather than after it.
- Standardized terminology and data-entry rules across institutions so sector reporting became comparable instead of being assembled from incompatible local interpretations.
- Modernized the live IBM Domino environment, migrated infrastructure from Windows to CentOS/nginx and strengthened authentication/access control without interrupting service.
- Restored public systems after false phishing/malware classification by coordinating remediation and delisting with Google, Bitdefender, Quttera and Sangfor.
- During the domain-block incident, normal domain access was impaired for roughly 1.5 days; maintained fallback access where possible and drove external escalation/recovery.
Deputy Head of Digital Development & IT / Acting Head of IT
Sep 2019 - Mar 2020
Joined a newly created organization where most end-user and operating IT had to be built quickly rather than inherited as a finished function.
- Helped stand up the IT function and workplace environment for approximately 100 users under an ASAP operating mandate.
- Built core end-user and operating IT from near-zero: workstations, domain and policies, network rollout/testing, printers and office hardware, software/licensing, support processes, server/directory integration and contractor coordination.
- Used standardized images, domain policies and parallel deployment methods to bring many workstations online quickly rather than configuring them one by one.
- Introduced HelpDesk and automated inventory to create operating visibility from the start.
- Planned departmental activity, controlled execution and participated in contractor selection and technical acceptance of major digital projects, including Healthy Moscow and My Social Center.
- Adapted an open-source internal mapping tool to solve a practical problem in a large multi-floor building where teams had difficulty locating departments and staff.
Lotus / IBM Domino Developer
Nov 2016 - Mar 2017
- Supported and developed an internal IBM Domino document-management environment used by 3,000+ users.
- Worked as third-line support for workflow and Domino application issues rather than only end-user support.
- Processed an inherited backlog of enhancement requests and improved consistency across internal Domino applications.
Lead Software Developer / Systems Architect
Jan 2013 - Sep 2015
The bank needed more than electronic storage. Document deadlines created regulatory and financial exposure, while physical routing made it hard to know where a document was and who had delayed it.
- Worked with executives and department heads to reconstruct document routes, ownership, deadlines, registration rules and regulatory constraints before automating them.
- When department heads disagreed about the process, escalated unresolved rules to executive governance, documented the agreed route in detail and only then implemented it.
- Designed and built an internal electronic document-management system from scratch on IBM Lotus Domino with XML-based 1C integration.
- Approximately 60 back-office users worked with the system.
- Replaced opaque paper movement with digital copies and explicit status/ownership so multiple employees could work with the same matter without physically circulating the original.
- After rollout, documents no longer disappeared inside the internal process; management could see location, owner, transfer time and overdue status and identify bottlenecks.
- Internal courier movement was no longer required for routine document circulation; the courier remained relevant for external delivery.
- The system addressed an environment where missed document/reporting deadlines created regulatory and financial exposure for the bank.
- Owned the path from process discovery through architecture, development, testing, rollout, training and support while coordinating another developer.
Lotus Notes Specialist
Jan 2011 - Apr 2012
- Developed and maintained IBM Domino / Lotus Notes workflow systems in a systems-integration environment.
- Introduced XPages into the company and rewrote the core Boss-Referent library in JavaScript / SSJS to support web interfaces.
- Modified Boss-Referent for web use, added modules/databases, fixed application issues and trained new employees on the internal workflow system.
Independent Technical Specialist
Jan 2009 - Jan 2011
- Built and maintained websites and Linux-based services, automated database updates and backups, and handled small-business infrastructure/support work.
- This period is where much of my practical Linux/web/infrastructure background developed before I moved deeper into enterprise workflow systems.
Junior Lotus Notes Developer
Sep 2008 - Sep 2009
- Selected through an internal training/hiring process from a group of 12 candidates.
- Developed, modernized, tested and supported Lotus Notes / Domino workflow applications in a large enterprise environment.
- Built service applications and agents, integrated Lotus applications with Microsoft Office through OLE, prepared documentation and worked with Service Desk/testers on production issues.
- End-to-end product work: discovery, requirements, process reconstruction, data model, workflow states, access rules, delivery, rollout, adoption and measurement.
- Multi-actor workflow design across clients, employees, institutions, departments and external partners.
- Product adoption judgment: willing to stop a technically capable product when user/organization fit is wrong rather than defend prior investment.
- Platform/product boundary judgment: distinguish what belongs in shared backend/platform capability from what belongs in product-specific UI or workflow.
- Stakeholder alignment: resolve contradictory process expectations before automating them.
- Scaled adoption: documentation, webinars, support and feedback loops for hundreds of users across 100+ institutions.
- Change management in resistant environments: introduced shared identity, structured HelpDesk use, centralized data, CRM/process systems and explicit operating rules despite initial user resistance.
- Business-process redesign and automation across finance, legal, procurement, service operations, customer/client workflows and regulated document processes.
- Architecture designed for change and handoff rather than dependence on the original implementer.
- Production Supabase backend operating for approximately three years through a complete frontend replacement.
- Production Xano system later able to replace the full backend from architectural documentation.
- Enterprise data modeling across complex many-to-many organizational structures and role-specific access.
- Legacy modernization in live environments, including IBM Domino, XPages and Windows-to-Linux/nginx migration.
- Banking/document workflows, XML/1C integration and enterprise third-line support at 3,000+ user scale.
- AI-assisted engineering through task decomposition, repository instructions, bounded context, acceptance criteria, guardrails, checkpoints and review rather than unstructured code generation.
- AIS MosRazvitie: sector-wide operational source of truth for 100+ cultural institutions and libraries, combining organizational data, event workflows, planning/reporting, RBAC, data-quality controls and scaled adoption.
- Exit Lead: field operations and client-reporting platform with approximately 70,000 historical documents migrated into a structured backend.
- Andronyevskaya ERP: self-hosted operational system with role-based workflows, Telegram Mini App access, QR-linked assets, tasks and automation.
- PromptlessPress: AI-assisted generation platform with reproducible prompt/execution pipelines, debugging interfaces and engineering guardrails.
- MetaFox Strengths Explorer: assessment platform with scoring logic, PDF reporting, peer feedback and administrative tooling.
I am not tied to one stack. Current projects usually combine an existing database, a web/low-code interface, APIs/webhooks, permissions, automation and some manual process that needs to be removed carefully. The tools I use most often now include PostgreSQL, Supabase, SQL, RLS, Edge Functions, WeWeb, Xano, n8n, Make, REST APIs, JavaScript/TypeScript, Python, Node.js, Docker, Git/GitHub, Telegram integrations and OpenAI APIs.
On the enterprise side I have worked with IBM/HCL Domino, Lotus Notes, XPages, LotusScript/SSJS, XML/1C integration, Active Directory, GPO, Windows Server, Linux, NGINX, VMware/Hyper-V, GLPI, monitoring, VPN, Synology, CCTV and access-control systems. That stack came from years of document workflow, internal systems, infrastructure and operational support rather than from a keyword checklist.
For AI-assisted delivery I use agents as managed executors rather than as a replacement for engineering control: explicit tasks, repository instructions, expected outcomes, schema/context boundaries, acceptance criteria, checkpoints, review and reproducible debug paths.
- Bachelor in Management, Small Business Management specialization — National Institute of Business, Moscow, 2019
- Professional retraining in IT for Economics and Government — Humanities Institute, Moscow, 2019
- Accounting and Finance background — Commercial Banking College No. 6, Moscow, 2006–2008
- Java SE fundamentals — Specialist Training Center, 2018
- ITIL / DevOps service-management training — Specialist Training Center, 2018
- IT management training — Specialist Training Center, 2018
- Zero-code / international product-building training — Zerocoder University, 2022
- PlanFix implementation courses — 2022
- n8n Certification — 2026
- Based in Belgrade, Serbia; work authorization in Serbia.
- Experience organizing events with 10,000+ participants.
- Built both public-facing products and internal operational systems.