KRISTERS GABRĀNS

Sales development and AI automation · Rauna, Latvia

Kristers Gabrāns

The things I've built and what they did for real people, plus proof of what's on my CV.

Contact

What I've built

My own product

ReviewTap

The problem
Happy customers rarely leave a Google review, and unhappy ones go straight to Google.
What I built
A tap card for the counter, the website and the dashboard behind it. Customers rate with stars first, so owners hear about problems privately and can fix them. Automated emails and a live count of taps and reviews.
Result
Sold the first 10 to businesses I didn't know, through cold messages.
Sales

My own outbound sales system

The problem
Finding the right people to contact takes hours, and tools like Clay are expensive.
What I built
Everything from scratch: email domains and inboxes, lead sourcing and qualification, and the copy. Plus my own cheaper Clay-style workflow that finds companies and enriches the data.
Result
2,623 agencies narrowed to 102 founders ready to contact, for about $6 in data.
Automation for a business

Accounting firm automations

The problem
A family-run accounting business's team spent hours on the same admin tasks every month.
What I built
Automations that take that repeat work off the team.
Result
About 15 hours saved every month.
Automation

Inbox Clerk

The problem
Orders and quotes arrive by email and someone has to type them into the accounting system.
What I built
A tool that reads the email, turns it into rows ready for the accounting system, and drafts the reply.
Result
The typing and the first reply are done before anyone opens the email.
Research

Latvia company scanner

The problem
A cold message that says nothing about the company gets ignored.
What I built
A scanner that reads Latvia's free open government data (tax records, the business register, annual accounts) and estimates where each company is losing money.
Result
3,779 Latvian companies mapped, so a first conversation can start from a number about their own business.
AI

AI agent teams

The problem
One AI working alone makes mistakes nobody catches.
What I built
Teams of AI agents that split up a job and check each other's work before anything goes out. They draw on a knowledge base I made from about 90 business books.
Result
Research and writing I can trust without redoing it myself.

Tools I use: Claude Code, MCP servers, APIs and webhooks, Supabase, Resend, Python, Zapier, Apify

Proof

SEB School Ambassador programmeSEB's screen inviting students to apply, with me as one of this year's mentors. Next to me is Linda Šablinska, the programme's director.
ViA Youth Innovation ForumCertificate No. 2025/40, Vidzeme University of Applied Sciences, 27 Feb 2025