AUTOMATION · AI INTEGRATIONS · DIGITAL SYSTEMS
Automation that survives real life.
I turn messy manual operations into reliable workflows and AI integrations, with the monitoring, fallback and secure handover real teams need.
Message me on WhatsAppopens in a new tabUse the contact form
Normally a reply the same business day.
See the 12-workflow onboarding system
Built by Dan Burdetsky, 15+ years across infrastructure, production operations and broadcast/streaming platforms, including VP of DevOps.
Proof
- 12 workflows. 8 integrations.
- Client onboarding for a professional services firm, across every service line.
- Deduplication, retries, alerts, fallback
- The reliability practices I brought from production operations, designed in rather than added after the first incident.
- Workflow tools plus custom components
- Python, JavaScript and API components for the parts a workflow tool cannot reach.
Manual work rarely lives in one place.
It sits across inboxes, WhatsApp chats, spreadsheets, task boards and the people who remember what happens next.
I map that process, connect the tools already doing useful work, and build the missing automation around the real edge cases.
- The same information is copied between tools.
- Documents and follow-ups depend on someone remembering.
- Nobody knows quickly when a workflow fails.
The same problems, in businesses that have nothing else in common.
I work by problem shape, not by industry. The vocabulary changes. The shape usually does not.
- The process grew with the business. Nobody designed it, and it is written down nowhere.
- It runs across tools that were each chosen for a different reason, by different people.
- Two teams each assume the other owns the handoff.
- It holds up until the volume doubles, someone is away, or a case arrives that nobody planned for.
None of that is specific to an industry. It is what happens when a process is assembled by people doing their jobs, rather than designed.
Three ways I can help
Business Automation
Connect intake, documents, tasks, customer updates and reporting into one visible process.
Explore business automation
AI Integrations
Use AI for the parts it handles well, with validation and a human review path around the parts it does not.
Explore AI integrations
Websites & Digital Systems
Build a professional, fast website that connects to the business behind it, not just a brochure that looks finished.
Explore websites
CLIENT ONBOARDING · CASE STUDY
Twelve workflows. Eight integrations. One tracked process.
A professional services firm was onboarding clients by hand across Slack, email, WhatsApp, Google Workspace and Trello.
I rebuilt the process on n8n Cloud. Documents are classified and routed automatically, customer updates are triggered at the right point, and unclear cases still go to a person.
- One error handler across the workflow suite.
- Deduplication across email and WhatsApp.
- Quota-safe writes to Google Sheets.
- Secure service-account access and daily backups.
The recovery path is part of the product.
A workflow that only works with clean inputs is a demo. Real operations include duplicate events, missing fields, slow APIs and people replying from the wrong address.
The same WhatsApp document arrived twice, two seconds apart, and got processed twice.
The same document could arrive by email and by WhatsApp. A Files Registry deduplicates across both channels, so the same document is never processed twice.
- Map the real process and its owners.
- Build the smallest useful workflow.
- Test messy inputs and failure modes.
- Add logging, alerts, fallback and secure access.
- Document the system and hand it over.
I build the automation and the reliability around it.
From 2021 to 2024 I was VP of DevOps for an interactive video and live-streaming platform, through its acquisition. I owned hybrid AWS and on-premises infrastructure, CI/CD, observability, disaster recovery and production reliability while leading a hands-on engineering team.
These were systems used by national broadcasters, where an outage is public, expensive and very hard to hide.
The platform was used by broadcasters including NBC News, MSNBC and ABC.
That background shapes how I build now. I expect things to fail, make the failure visible, and design a safe way back.
Notes from real automation work.
The best AI automation habits I have came from DevOps, not AI
Idempotency, one shared error handler, scoped credentials, alerts and a human review path. The habits that keep an AI automation alive are not AI habits.
A workflow that only works with clean inputs is a demo
Duplicate events, missing fields and slow APIs are normal production conditions, not edge cases. Three real failures and what they changed about the build.
What is your team still doing by hand?
Tell me where the process is slow, repetitive or easy to drop. I will tell you whether automation is likely to help and what I would look at first.
Message me on WhatsAppopens in a new tabUse the contact form
Normally a reply the same business day.