Skip to content
AlexDevCoreAlexDevCore

AI Automation & AI Agents Engineer

I build AI automations that do your repetitive work.

For small businesses and busy teams: AI agents, workflow automation, and clean API integrations that cut manual work, speed up client replies, and keep leads from slipping through the cracks.

I make the boring parts run themselves.

See selected work

Free audit · No obligation · You keep everything I build.

  • Less manual work
  • Faster client responses
  • Fewer lost leads
workflowruns 24/7
  • Trigger

    New lead arrives

    form · email · DM

  • AI Agent

    Qualify & enrich

    reads, decides, drafts

  • Actions

    Update CRM · Reply

    logged & followed up

The full process

How we'll work together

A simple, low-risk path from “this is annoying” to “this is handled.”

The Automation Sprint · 5 steps
  1. Step 1

    Discover

    A quick back-and-forth over text to understand your process and where the time actually goes.

    Your process, mapped
  2. Step 2

    Design

    I map the solution and pick the right tools for your stack and budget.

    A clear plan
  3. Step 3

    Build

    You get a working prototype fast, so you can see the value before committing to the full build.

    Working prototype
  4. Step 4

    Test

    I run it against real cases and smooth out the edges before it goes live.

    Tested on real cases
  5. Step 5

    Handoff & improve

    I deliver it documented and easy to use, then refine it as you put it to work.

    Documented & live

Skills & tools

My stack

The tools I actually ship with — plus the ones I'm leveling up on right now.

AI & LLMs

  • Anthropic Claude API
  • OpenAI API
  • Prompt engineering
  • AI agents
  • LLM app design

Automation & integration

  • Python
  • REST APIs & webhooks
  • n8n
  • Make / Zapier
  • Cron / scheduling
  • Workflow design

Data & web

  • Web scraping
  • Data extraction & cleaning
  • CSV / Excel pipelines
  • Flask
  • Next.js / React
  • Tailwind CSS
  • Multilingual / i18n

Engineering & tooling

  • Git & GitHub
  • Docker
  • TypeScript / Node.js
  • Linux / Ubuntu
  • WSL
  • Bash
  • PostgreSQL / SQL
  • Testing (pytest)
  • Railway (deploy)
  • VS Code

The problems I fix

If any of these sound familiar, they can be automated.

Manual, repetitive work

Copy-pasting between apps, re-typing the same data, sending the same replies by hand.

See examples
  • Copying new orders from email into a spreadsheet by hand
  • Re-entering the same client details into three different tools
  • Sending the same “got your request” reply over and over

Slow client responses

Inquiries sit unanswered for hours — especially after business hours — and interest cools off.

See examples
  • A quote request lands at 8pm and isn’t answered until noon
  • Weekend inquiries pile up until Monday morning
  • Leads go cold while they wait hours for a first reply

Leads slipping through

Messages arrive across emails, forms, DMs, and some never get followed up.

See examples
  • A contact-form submission nobody noticed for a week
  • A DM that got buried and never followed up
  • No single list of who reached out and where they stand

Tools that don't talk

Your CRM, spreadsheets, inbox, and other apps each hold a piece, but none of them connect.

See examples
  • Your CRM has no idea about orders sitting in a spreadsheet
  • New leads in one app never reach your email tool
  • You export and re-import CSVs by hand to move data around

Data scattered everywhere

The numbers you need live in five places, so every report is a manual chore.

See examples
  • Numbers split across Stripe, a spreadsheet, and your inbox
  • Building one weekly report means copying from four places
  • No single view of how the business is actually doing

AI you can't put to use

You know AI could help somewhere — it's just not clear where it actually pays off.

See examples
  • You’ve tried ChatGPT but can’t wire it into your workflow
  • Not sure which tasks are safe to hand to AI
  • You want something that acts on your data, not just chats

Services

Core focus areas

Practical automation, built around how you actually work — not a template.

01

AI Automation

End-to-end automation of repetitive processes, so routine work runs itself instead of eating your day.

For example
  • Auto-file every incoming order into your spreadsheet and CRM
  • Turn one filled form into a task, an email, and a calendar event
  • Nightly jobs that prep your reports before you wake up
02

AI Agents

Assistants that read your data, answer questions, and take action — not just chat back at you.

For example
  • A support assistant that answers FAQs from your own docs
  • An agent that reads each new lead and drafts a tailored reply
  • A research agent that summarizes and tags incoming info
03

Workflow Automation

Connect your apps so work flows automatically, with no copy-paste. Built with n8n, Make, or custom code.

For example
  • Form → CRM → email → Slack, all in one automatic flow
  • Keep two apps that don't natively talk perfectly in sync
  • Fire off a whole sequence from a single event or click
04

API Integrations

Make your tools talk to each other — CRM, email, spreadsheets, webhooks, and the apps you already use.

For example
  • Push new Stripe payments straight into your accounting sheet
  • Sync contacts between your CRM and your email tool
  • Wire up a webhook so one app reacts to another instantly
05

Business Process Automation

Automate lead capture, client intake, follow-ups, and reporting, so nothing depends on someone remembering.

For example
  • Capture leads and send an instant, personal follow-up
  • Client onboarding that runs itself, step by step
  • Scheduled reports delivered to your inbox on time, every time
06

AI Consultation & Process Audit

Not sure where to start? I'll map your process and find the highest-impact thing to automate first.

For example
  • A clear map of your process with the top 3 things to automate
  • An honest estimate of the time and money you'd save
  • A no-jargon plan you can act on — even without me

Previous projects

Focused, client-oriented projects — each one solving a real problem end to end.

AI / LLM application

Document Explainer

Explains confusing official US letters in plain language — and in your language.

  • Python
  • Flask
  • Anthropic Claude (claude-sonnet-5)
  • pytest
CodeLive demo

The problem

Official letters — from the DMV, IRS, a bank, a landlord, or an insurer — are full of jargon and deadlines. People who are new to the US often can't tell an urgent notice from junk mail, and end up relying on second-hand interpretation.

Who it's for

People new to the US and the systems around them — and honestly, anyone who wants a dense official letter turned into something they can act on in 30 seconds.

What it does

A web app where you paste the text — or upload a photo or PDF — of a letter and get four clear things back: what it is, what you need to do (with any deadlines), what happens if you ignore it, and the whole summary translated into your language.

Key decisions

  • Forced the AI output into four fixed sections, so the answer is always scannable and actionable — never a wall of text.
  • Added a clearly-marked demo mode that runs with no API key, so the interface can be tested instantly and safely.
  • Built in translation and a multilingual UI, so the tool meets people in their own language.
  • Kept a firm boundary in the copy: it explains what a document says — it is not legal, medical, or financial advice.

What I learned

How to turn an LLM into a dependable product feature instead of a demo — shaping the prompt for consistent structure, handling the no-key case gracefully, and writing tests around AI output.

Where it goes next

Save explanation history, support more document types, and add a mobile-friendly camera-capture flow.

Web scraping · Data automation

Catalog Scraper

Turns any paginated catalog site into a clean, ready-to-use spreadsheet.

  • Python
  • requests
  • BeautifulSoup
  • openpyxl
Code

The problem

The data a business needs — competitor prices, product catalogs, lead lists — is often trapped across dozens of web pages. Collecting it by hand is slow, boring, and easy to get wrong.

Who it's for

Anyone who needs web data as a spreadsheet: market research, price monitoring, building a catalog, or assembling a lead list.

What it does

A Python scraper that crawls a paginated catalog site, extracts structured fields for each item (title, price, stock, rating, detail link), follows pagination automatically, and exports clean CSV and Excel files ready to open in Excel or Google Sheets.

Key decisions

  • Built it as a reusable skeleton, not a one-off script — you swap the target URL and a few CSS selectors, and the pagination, export, and rate-limiting logic stays the same.
  • Polite by design: a custom User-Agent and a small delay between requests, so it stays respectful of the target site.
  • Exports to both CSV and Excel with auto-sized columns, so the output is usable the moment it lands.

What I learned

How to design a scraper that's reusable across sites, and how small politeness details — headers, delays — are what separate a throwaway script from something you can actually run for a client.

Where it goes next

Move selectors into a config file, add scheduled runs, and push results straight to Google Sheets or a database.

The demo runs against books.toscrape.com, a public sandbox built specifically for scraping practice.

About Me

I'm Aliaksei — I go by Alex — an AI automation engineer based in San Diego. I studied Artificial Intelligence at University and have spent years since building things and running hands-on, operations-heavy work.

That mix is the point. I don't just write code — I understand the messy day-to-day of a small business: the missed messages, the manual follow-ups, the data spread across five different tools. So I build automations around the real problem, not a generic template.

When you work with me, you work directly with me. Fast replies, real ownership, and a solution shaped around how you actually operate.

Aliaksei Krachkouski

AI Automation & AI Agents Engineer

Available for new projects
Based in
San Diego, CA
Working
Remote · Worldwide
Response
~10 minutes
Focus
AI automation & agents

Frequently asked questions

What kind of businesses do you work with?

Mostly small and mid-sized businesses, solo operators, and teams I meet through platforms like Upwork — agencies, e-commerce shops, service providers, real-estate and local businesses, and anyone buried in repetitive admin. You don't need to be technical or have a big budget. If part of your day is spent copying data between apps, chasing leads, or doing the same task over and over, that's exactly the kind of work I take off your plate — and hand that time back to you.

What can actually be automated in my business?

Far more than most people expect. The simple test: if a task follows rules and repeats, it can usually be automated. Common ones I build: lead capture with instant follow-up, client intake and onboarding, appointment scheduling, quote and invoice generation, data entry and CRM updates, report generation, moving and syncing data between apps, chatbots that answer repetitive questions from your own docs, parsing PDFs and documents, order and payment processing, review and feedback requests, and scraping or cleaning data from the web. Not sure whether your specific task qualifies? Send it over — I'll tell you honestly whether it's worth automating, and what it would save you.

How much does it cost, and how long does it take?

Every project is a fixed quote agreed up front after a free audit — no hourly surprises, no scope creep. As a rough guide, based on the work I actually deliver: a quick single-workflow automation (e.g. form → CRM → email) usually takes a few days; a standard integration syncing 2–3 tools with follow-ups and notifications takes about 1–2 weeks; an AI agent or chatbot that answers from your docs or drafts replies runs around 2–3 weeks; a full multi-step process (intake → onboarding → reporting) is roughly 3–5 weeks; and a larger custom system with several agents or a complete ops pipeline is 6+ weeks, scoped in phases so you see value early. Real timelines depend mostly on how fast we can access your tools and test on live data — you'll always get a concrete estimate before anything starts.

Can you build a prototype before I commit?

Yes, and I recommend it. I usually start with a working prototype of the highest-impact piece, running on your real data, so you can see it actually work before committing to the full build. You tell me what to adjust, and you only move forward once you genuinely love it. It keeps your risk low — you're never paying for something you haven't already watched deliver.

How does a project start?

Just send me a short message about what's slowing you down — no formal brief needed. I'll ask a few quick questions over text, dig into your process, and pinpoint the one or two things worth automating first. Then I come back with a clear plan, a timeline, and a fixed price, so you know exactly what you're getting before any work begins.

Can you integrate with the tools I already use?

Almost always — that's the core of the work. I connect CRMs, spreadsheets, email, calendars, payment tools, and just about anything with an API or webhook, so your team keeps the tools it already knows. No rip-and-replace. If a tool has no official integration, I can usually still wire it up with custom code.

Do I need any technical knowledge?

None at all. I handle everything technical and hand you something you can actually use, with plain-language documentation and a short walkthrough. Most of my clients aren't technical — you describe the problem in normal words, and I turn it into something that quietly runs itself in the background.

Free resource

Not ready to talk yet?

Grab my free Automation Starter Checklist — the 12 things worth automating first in a small business, plus a one-page way to work out which one is costing you the most. The PDF downloads the moment you hit send.

No spam. I'll only email you if you ask me to.

Have a process that's eating your time?

Let's find the highest-impact thing to automate first. Book a free, no-pressure audit — I'll show you where AI actually helps, with no jargon.

Email me

Free audit · No obligation · You keep everything I build.