Product Engineer & Independent Software Builder

I turn business problems and unfinished ideas into working software.

I build SaaS products, automate operations, rescue unreliable applications and publish open-source tools.

Discuss a project See how I work

Or read the code on GitHub

7+ years of engineering · Products I own and ship · Open source you can read

ownership.ts

Two ways to work with me

Hire me, or bring me a problem

Both end in the same place: one engineer who owns the problem from definition to production.

01 / Hire me

A role on your team

Senior Product Engineer, Founding Engineer or Tech Lead, full-time or long-term contract. Best fit when the work needs someone who decides what to build, not only how to build it.

See my background Start a conversation

02 / Bring me a problem

A product, a process or a system that stalled

Work that starts from a situation rather than a specification. You describe what is stuck; I work out what should be built and take it to production. The three directions below are what that usually turns out to be.

Describe the situation

Build · Automate · Rescue

What bringing me a problem looks like

Different problems, the same way of working: understand the problem, define the smallest useful system, own it through to production.

01 / Build

Build Products

I design and build SaaS products, APIs, internal systems and web applications, from product definition to production.

Explore products

02 / Automate

Automate Operations

I replace repetitive business work with systems that connect tools, capture requests and keep processes moving.

Explore automation

03 / Rescue

Rescue Software

I investigate, stabilize and finish applications built by AI, freelancers or previous development teams.

Rescue a project

Proof

Products, tools and systems I'm building

Everything above is a claim. This is the part you can check: shipped projects with real versions, published source and current status, including what is not finished.

Open Source · Self-hosted

AlertLoop

Released

A self-hosted event and notification center for software teams.

Problem: Important incidents and business events disappear across logs, inboxes and disconnected channels.
Result: AlertLoop collects events, tracks incidents from firing to resolved, routes notifications by rule and keeps delivery failures visible and replayable.
My role: Product definition, architecture, implementation direction and release management.
Status: Community edition released as an alpha; routing rules, incident lifecycle and recovery notices included. Pro and Enterprise editions are planned, not built.

View project GitHub

Open Source · Developer Tool

Traceloom

Released

Structured JSONL event timelines, grouped by trace_id, without a heavy framework.

Problem: Simple logs often lose the context needed to understand what actually happened inside an application.
Result: Traceloom records structured execution traces without forcing a large framework or an external platform.
My role: API design, architecture, documentation and cross-language product strategy.
Status: PHP, JavaScript and Go implementations released. One event format across all three.

View project GitHub

Commercial Product

SaaS for Service Businesses

In Validation

A product designed to help small service businesses in Latin America stop losing WhatsApp bookings while they are busy with clients.

Problem: Booking requests arrive exactly when the owner cannot answer immediately.
Result: The product is designed around recovered time, fewer missed bookings and reduced no-shows, not around the concept of a bot.
My role: Product, architecture and implementation.
Status: In validation. A working demo is being shown to potential clients; details stay private until the product is public.

Get in touch Read the build notes

Open Source

Libraries and tools you can read

Every open-source project above is published under a real license, with a changelog and tagged releases. Weft, the PHP engine behind this site, is part of it and runs this page in production.

View open source

More is in progress

I'm developing additional SaaS products, developer tools and automation systems. New projects will appear here as they reach a useful public stage.

See everything, including what's planned

How I work

From unclear problem to production

Most projects don't fail on syntax. They fail on the wrong problem, an unclear boundary or a system nobody owns after launch.

Step 01

Define the real problem

I separate the actual business or engineering problem from the first proposed solution.

Step 02

Design the smallest viable system

I define the product boundary, architecture and validation path before scaling implementation.

Step 03

Own the decisions, direct the agents

Architecture, technical direction and review stay with me. AI agents increase delivery speed inside those boundaries.

Step 04

Test, ship and observe production

A feature is not finished when code is generated. It is finished when it works under real conditions.

Step 05

Turn the result into reusable knowledge

I document decisions, publish useful tools and convert lessons into stronger products.

AI increases my production capacity. It does not replace engineering ownership.
Roman Golovanov

Notes

Engineering and product notes

Decisions, tradeoffs and build logs, written while the work is still fresh.

AI Workflow·July 2026·9 min

How I Use AI Agents Without Giving Up Engineering Ownership

Product, architecture, release and publication decisions stay with me, while AI agents write most of the code. The working system: fact statuses with sources, session gates, and two failures that shaped the rules.

Build Notes·July 2026·10 min·Traceloom

Building one library across PHP, JavaScript and Go

Traceloom exists in PHP, JavaScript and Go: three repositories, three version lines, one JSONL record format. What made the port work was one decision: the file on disk is the product, and everything above it is a local dialect.

Build Notes·July 2026·6 min·Weft

Why Weft is deliberately not a framework

Weft ships with a rejection list instead of a roadmap: no ORM, no DI container, no admin panel. What that decision buys a content-first site engine, and what it costs.

Product Decisions·July 2026·6 min·AlertLoop

Why I'm building AlertLoop as a self-hosted product

Cloud was the faster path to revenue. I chose a self-hosted Community Edition first: the reasoning, the options I rejected and the tradeoffs I accepted.

View all notes

About

Product thinking. Engineering ownership. Real delivery.

I'm a product-minded software engineer with 7+ years of hands-on development experience. My work spans backend systems, APIs, web applications, integrations, automation and product architecture.

I'm most useful when a project needs someone who can understand the problem, make technical decisions and move the product forward without a large team.

Today I build my own products, publish open-source tools and help companies create or stabilize software systems.

Open to Senior Product Engineer, Founding Engineer and Technical Lead roles

Have a product to build, a process to automate, or software that needs rescuing?

Send a short description of the situation. I'll tell you whether I can help and what the next useful step should be. Hiring instead? This form or LinkedIn, whichever you already have open.

Describe your situation Hiring? Start on LinkedIn

JOB Hiring for a role? Senior Product Engineer · Founding Engineer · Tech Lead GH Reviewing my code first? github.com/golovanov-dev