Oleksandr Shylo
Offer Oleksandr work on your next project.
Private message 
Offer work 
Kharkiv,
Ukraine
currently online
available for hire
8 proposals made
on the service 14 years
Rating
Language proficiency level
Українська: fluent
Русский: fluent
English: upper-intermediate
Skills and abilities
Programming
Design & art
Photo, Audio & Video
Promotion
Outsourcing & consulting
Portfolio
- All works 11
- Lead Generation & Sales 5
- Data Parsing 1
- AI & Machine Learning 1
- Bot Development 3
- Hybrid Mobile Apps 1
-
891 USD Mobile App for Clinics — No App Store, No Google Play
Hybrid Mobile Apps
Your patients can book appointments directly from their phones — no downloading anything from a store. They scan a QR code at reception or tap a link in Instagram, and the app is on their home screen.
What's included in the base version:
— Online booking: choose service, doctor, date & time
… — Doctor profiles with photos, specialty, and patient reviews
— Contacts, map, WhatsApp button, callback request
— Works offline
— Updates automatically — patients never need to reinstall
Who it's for: dental clinics, aesthetic medicine, massage studios, beauty salons, private practitioners.
Why not just a website: the app lives on the phone's home screen as an icon. Patients open it with one tap — no address bar, no browser friction. Higher conversion to bookings.
Why not a native app ($10,000+): PWA technology delivers 90% of native app functionality at 10% of the cost. No Apple/Google developer account needed, no review delays, no 30% commission.
Stack: HTML/CSS/JS · PWA · Cloudflare (hosting, HTTPS, CDN)
Timeline: base template — 3–5 days, custom design — 1–2 weeks
Live demo: https://artsmile-demo.pages.dev
-
156 USD Telegram Prediction Bot for FIFA World Cup 2026
Bot Development
A fully functional Telegram prediction bot built and deployed for a private group tournament running alongside the 2026 FIFA World Cup.
What was built:
… The bot automatically tracks all tournament matches, collects predictions from participants, calculates points, and posts results to the group chat immediately after each match ends. Participants place two types of bets per match — on the winner (1 point) and on the exact score (3 points).
Tech stack:
Python 3.12 + aiogram 3.x — async Telegram bot with inline keyboards, callback handlers, and a full state management system
aiosqlite — async SQLite for matches, bets, users, and a full transaction ledger
Docker + DigitalOcean VPS — containerized deployment on an Ubuntu server with automatic restart on failure
football-data.org API — official source for the full match schedule, statuses, and final scores (104 matches)
ESPN Unofficial API — secondary real-time data source polling live scores every 2 minutes during active matches, no API key required
Make.com Webhook → Google Sheets — after each settled match, all bets and results are automatically logged to a spreadsheet
Dual-source score sync — ESPN updates the score first (priority); football-data.org only updates match status without overwriting live score data
Key features:
Bet placement and cancellation with a 10-minute change window
Bet hiding until the change window expires (copy-protection)
Automatic points calculation and group announcement on match end
Real-time leaderboard
Entry fee payment confirmation system with personal and group notifications
Admin command to attach stream links to upcoming matches
Atomic race-condition protection preventing duplicate match settlement
USDT TRC-20, Aptos, and Bybit payment method support
-
500 USD Automation of bookings and cancellations — Cal, Pipedrive, AC
AI & Machine Learning
Built a complete automation for bookings and cancellations through Make, Cal, Pipedrive, and ActiveCampaign. When a client books a discovery call through Cal, Make automatically searches for the contact in Pipedrive, routes through the deal history, creates or updates the contact and deal, logs it in Google Sheets, and triggers the necessary email sequence in ActiveCampaign. When the call is canceled, a separate branch updates the deal stage in Pipedrive, tags the contact in ActiveCampaign, and launches a re-engagement sequence. Each booking and cancellation is processed automatically, with zero manual work.
Built an automated AI content generation system using Make and OpenAI. Telegram triggers the workflow — Make routes content through four parallel branches: photo posts, text and image posts, audio.
-
350 USD Automation of lead capture and nurture chain for ebook
Lead Generation & Sales
Built an automated lead capture and nurture system for ebook opt-ins. When a new subscriber fills out a ClickFunnels form, Make triggers automatically — searches Pipedrive by phone and email, checks for duplicates, routes the lead through qualification logic. New leads get a person and deal created in Pipedrive with full activity logging. ActiveCampaign tags the contact, adds to the right list, and launches a nurture email sequence. All data logged to Google Sheets. Zero manual entry from opt-in to first follow-up email.
Make, Pipedrive, ActiveCampaign, ClickFunnels, Email Marketing, Lead Generation
-
400 USD AI lead qualification and CRM automation
Lead Generation & Sales
Built a multi-level lead qualification automation through Make, Pipedrive, and ActiveCampaign. A new lead comes in from a form — Make checks Pipedrive for duplicates, qualifies based on source and data, creates deals, adds activities and notes, logs to Google Sheets, and triggers the necessary email sequence in ActiveCampaign. Qualified and unqualified leads receive different follow-up paths. Zero manual processing, no leads are lost.
Make, Pipedrive, ActiveCampaign, CRM, Lead Generation, AI Automation
-
67 USD Web Scraping and Lead Generation | Local Business Data
Data Parsing
Built automated systems for collecting data on local businesses in several niches — dentistry, fitness studios, beauty salons, and makeup artists. Collected company names, addresses, phone numbers, emails, websites, ratings, and reviews. Data cleaned, deduplicated, and formatted into structured Google Sheets, ready for cold outreach or CRM import. Each scraping process is set to automatically run on a schedule to gather fresh data. Used for cold outreach campaigns and lead generation funnels connected to Pipedrive and ActiveCampaign.
-
178 USD AI automation: portfolio website with integration Make.com → Telegr
Lead Generation & Sales
https://makeflows.netlify.app/
Developed a complete client acquisition system from scratch.
What was done:
…
— Designed and built a one-page website for an AI automation specialist. The site includes an interactive ROI calculator, an animated data flow diagram, a services section, and a contact form.
— Connected the form to Make.com via Webhook. When the form is filled out by a client, the system automatically:
- adds the request to Google Sheets with all client data
- sends a structured message to a Telegram group
- sends a personal email auto-response to the client
— Deployed the site on Netlify. The system operates 24/7 without manual intervention.
Tools: Make.com · Webhook · Google Sheets · Telegram Bot API · Gmail · HTML/CSS/JS · Netlify
-
122 USD AI Content Factory: Instagram posts in a minute
Bot Development
AI CONTENT MACHINE FOR INSTAGRAM
Author's system for automatic content generation with multimodal input. From idea to finished post — in 30 seconds.
… 4 INPUT METHODS:
• Text message
• Voice message (Whisper)
• Photograph (GPT-4 Vision)
• Bulk generation of 5 posts
HOW IT WORKS:
Topic in Telegram → GPT-4 writes storytelling in English → DALL-E 3 creates images → finished post in your hands.
STACK:
Make.com, GPT-4, DALL-E 3, Whisper, GPT-4 Vision,
Telegram Bot API, Google Drive, Google Sheets
RESULT:
30 seconds per post (was 4 hours)
35 posts/week (was 7)
$0.40 per post (was $50)
95% time savings
-
600 USD Chatbot for a dental clinic: auto-scheduling + CRM integration
Bot Development
CLIENT: Private dental clinic (Kyiv, 4 doctors)
PROBLEM:
The administrator was unable to process requests from Instagram and the website, especially in the evenings and on weekends. Some patients got lost — they wrote to several clinics at the same time, choosing those who responded first. There were also many no-shows (~20% of appointments).
…
SOLUTION:
AI chatbot 24/7 with automatic appointment scheduling, doctor selection, and reminder system.
FUNCTIONALITY:
— GPT-4 dialogue in Ukrainian
— Clarification of the patient's problem (pain, aesthetics, prevention, pediatric dentistry)
— Doctor selection by specialization
— Real-time availability check
— Appointment recording in the clinic's CRM with confirmation
— SMS + Telegram reminders 24 hours and 2 hours before
— Collection of feedback after the appointment
— Transfer of non-standard requests to the administrator
STACK:
n8n, OpenAI GPT-4, Telegram Bot API, Instagram Direct API, Google Calendar, TurboSMS, integration with the clinic's CRM
RESULTS FOR THE FIRST MONTH OF WORK:
✓ Processed 240+ inquiries (the bot closed ~70% automatically)
✓ Scheduled appointments: 89 new patients
✓ Response time reduced from 1-3 hours to 30 seconds
✓ No-show rate decreased from 20% to 7% due to reminders
✓ The administrator was freed from routine tasks — now only handles complex cases and consultations in the clinic
✓ Collected 24 new reviews on Google Maps
DEVELOPMENT TIME: 16 days
SUPPORT: 30 days free + optional follow-up
P.S. Specific financial indicators are not published due to NDA, but the client was satisfied and ordered a second bot for the new branch opening.
-
1100 USD AI qualification of leads for a business coach: $180K in 2 months
Lead Generation & Sales
ABOUT THE PROJECT:
A business coach from the USA sells group programs for scaling businesses priced between 5,000 and 15,000 USD. The website had a lead application form that brought in about 100 leads per week, but the sales manager spent hours manually filtering them — most applications were from people without a budget, not from the target audience, or just curious. There were only 8-10 real quality leads per week.
… TASK:
- Automate lead qualification based on 6 criteria
- Filter out low-quality applications before they reach the manager
- Automatically schedule calls for suitable leads
- Integrate the system with the existing CRM (HubSpot)
- Create internal analytics for lead effectiveness
SOLUTION:
Built an AI pre-qualification system:
STEP 1. The lead fills out a form on the Webflow site (12 questions: business size, annual income, niche, pain points, budget, willingness to invest).
STEP 2. Make.com receives the data and sends it to GPT-4 with a custom prompt that considers the criteria of the coach's ideal client.
STEP 3. The AI evaluates the lead based on 6 parameters and assigns a final score from 1 to 100, adding a comment on why this program is suitable or not for the lead.
STEP 4. HIGH SCORE (75-100): the lead automatically receives a link to Calendly with available slots of the top manager, data is stored in HubSpot with the tag Premium Lead, and a notification is sent to the team via Slack.
STEP 5. MEDIUM SCORE (40-74): the lead receives an invitation to a free webinar or mini-course for warming up, and is added to a 14-day email sequence.
STEP 6. LOW SCORE (1-39): the lead receives a polite rejection with a recommendation for free resources. In the CRM, it is stored with a tag for future checks.
STEP 7. Once a week, an automatic report with analytics arrives in Notion: how many leads, which traffic sources provide the best leads, average score, conversion by segments.
TECH STACK:
Webflow, Make.com, OpenAI GPT-4, Calendly, HubSpot CRM, Slack, Notion API.
RESULTS AFTER 60 DAYS:
- 180,000 USD revenue from qualified leads
- 87% accuracy of AI qualification (confirmed by the manager)
- 24 scheduled calls per week (up from 8-10)
- Sales conversion increased by 4.2 times
- Saved 25 hours of manager's work per week
- Payback period for the system: 11 days
WHAT THE CLIENT RECEIVED:
- A fully functional AI qualification system operating 24/7
- A custom prompt tailored to their ideal client
- Integration with all existing tools
- Weekly analytics in Notion
- Documentation and video instructions in English
- 30 days of free support after launch
Do you want a similar system for your consulting or coaching practice? Write to me — we will discuss your task and calculate the cost based on your goals.
Cost starting from 1,200 USD (depends on the complexity of qualification criteria and the number of integrations).
Implementation time: 10-14 days.
-
553 USD How an online English school received 287 applications in 3 weeks
Lead Generation & Sales
ABOUT THE PROJECT:
An online English language school with an audience in Ukraine. They launched ads on Facebook and Instagram for a free consultation but were losing up to 40% of leads—managers couldn't process requests during peak hours, and cold leads mixed with hot ones.
… TASKS:
- Automate the collection and qualification of leads
- Reduce response time from 2-3 hours to instant
- Separate hot and cold leads for managers
- Nurture those who are not ready to buy immediately
- Free managers from routine processing of junk requests
SOLUTION:
Built a complete lead generation funnel with AI qualification:
STEP 1. The user goes from the ad to a Typeform quiz (7 questions: language level, learning goals, budget, readiness to start).
STEP 2. Make.com receives a webhook and passes the responses to GPT-4.
STEP 3. AI analyzes the responses and assigns a lead temperature from 1 to 10, as well as generates a personalized conversation script for the manager.
STEP 4. HOT LEADS (rating 8-10): a deal is instantly created in Pipedrive + the manager receives a Telegram notification with a complete brief and a ready call script.
STEP 5. WARM LEADS (rating 5-7): automatically enter ActiveCampaign for a 7-day email sequence with alumni case studies and an invitation to a free trial lesson.
STEP 6. COLD LEADS (rating 1-4): added to a long nurture scenario with useful content about language learning, without aggressive sales.
TECH STACK:
Facebook Ads, Instagram Ads, Typeform, Make.com, OpenAI GPT-4, Pipedrive, ActiveCampaign, Telegram Bot API.
RESULTS IN 3 WEEKS:
- 287 qualified applications (previously 180 in the same period)
- Response time for a hot lead: 5 seconds (was 2-3 hours)
- Conversion to payment
