You will get a custom AI agent or OpenAI API script developed
Rising Talent
Rising Talent
Project details
I build custom AI tools and agents using Claude and OpenAI APIs. I develop Python scripts, bots, and automation pipelines for tasks that require human judgment.
What you get:
Working Python tool or AI script with full source code.
Smart integration: Claude or OpenAI chosen based on your task.
Advanced features: Prompt engineering, memory, structured JSON output, and tool use.
Clean, documented code you can easily extend or modify.
Common use cases:
Data Processing: Converting unstructured text or PDFs into structured data.
Text Processing: Document summarization and classification pipelines.
Smart Assistants: Customer support agents and custom AI bots.
Generation: AI email drafting and automated code review bots.
I work asynchronously and communicate in writing. You must provide your own API keys for testing.
Please message me before placing an order to discuss your project requirements first.
What you get:
Working Python tool or AI script with full source code.
Smart integration: Claude or OpenAI chosen based on your task.
Advanced features: Prompt engineering, memory, structured JSON output, and tool use.
Clean, documented code you can easily extend or modify.
Common use cases:
Data Processing: Converting unstructured text or PDFs into structured data.
Text Processing: Document summarization and classification pipelines.
Smart Assistants: Customer support agents and custom AI bots.
Generation: AI email drafting and automated code review bots.
I work asynchronously and communicate in writing. You must provide your own API keys for testing.
Please message me before placing an order to discuss your project requirements first.
AI Algorithms
Large Language Model, Multimodal Large Language Model, Transformer ModelAI Applications
AI Chatbot, AI Content Creation, AI Mobile App Development, AI-Generated Code, Conversational AI, Natural Language Generation, Natural Language Understanding, Sentiment Analysis, Synthetic Data GenerationAI Development Language
PythonAI Tools
Azure OpenAI, PyTorchAI Models
ChatGPT, GPT-4, LLaMAWhat's included
| Service Tiers |
Starter
$90
|
Standard
$160
|
Advanced
$280
|
|---|---|---|---|
| Delivery Time | 3 days | 5 days | 9 days |
Number of Revisions | 1 | 2 | 3 |
AI Model Integration | |||
Batch Normalization | - | - | - |
Database Integration | - | ||
Detailed Code Comments | |||
Image Upscaling | - | - | - |
MLOps | - | - | - |
Model Deployment | - | - | - |
Model Documentation | - | - | - |
Model Monitoring | - | - | - |
Model Testing & Optimization | - | - | - |
Model Tuning | - | - | - |
Natural Language Processing | |||
NLP Tokenization | - | - | - |
Pre-Training | - | - | - |
Prompt Engineering | |||
Setup File | |||
Source Code |
Optional add-ons
You can add these on the next page.
Fast Delivery
+$20 - $70
Slack or Telegram Bot Integration
+$50Frequently asked questions
2 reviews
(2)
(0)
(0)
(0)
(0)
This project doesn't have any reviews.
MB
Mary Jane B.
Jun 29, 2026
Typeform Logic Expert
Excellent!
CB
Catherine B.
Jun 18, 2026
Chrome Extension for Grok to Blogger
I'm very satisfied. Adam created a user friendly Chrome Extension saving me countless hours of work.
About Adam
Web Scraping | Chrome Extensions | WordPress Speed | Web Design
100%
Job Success
Lodz, Poland - 12:58 am local time
Recent work: three live business sites, hand-coded with no page builder, scoring 97 to 100 on Google PageSpeed on mobile, a multi-page salon site, a service landing page with a live price calculator, and a SaaS product landing. A public WordPress speed case study too: the same landscaping page built three ways, 63 on a bloated Elementor build, 71 optimized in place, 99 on a lean rebuild, every trade-off measured. Excel-driven scrapers for a client where targets are read from a spreadsheet and results written back, across LinkedIn, USPTO, WIPO, and G2. A Chrome extension that moves AI-generated content straight into Blogger posts, delivered and in use. A price monitor where you paste a product photo, Google Vision turns it into the search query, and prices from five marketplaces come back side by side, running 24/7 in Docker.
A typical build for me: a scraper watches a site, Claude decides what matters, a bot sends the result to your phone. For social platforms I stick to official APIs (Telegram Bot API, Discord, WhatsApp via Twilio, Instagram and Facebook via Meta Graph API). If a job needs mass DMs or fake engagement, I pass. That gets accounts banned.
What I take on:
- Web scraping and price monitoring (httpx, Playwright, BeautifulSoup4)
- Chrome extensions (Manifest V3, OAuth2)
- Fast websites and landing pages (hand-coded HTML/CSS/JS, no page-builder bloat, green on mobile PageSpeed, per-page technical SEO)
- WordPress speed optimization: measure the site, fix what a free pass can fix (images, render-blocking CSS, lazy-load, plugin and font cleanup), and say when a lean rebuild beats paying for more plugins
- PDF-to-Excel pipelines (pdfplumber, openpyxl)
- Telegram and Discord bots: alerts, lead capture, group admin, content posting
- n8n workflows with LLM steps (Claude, GPT)
- API integrations: Gmail, Google Sheets, Google Vision, Twilio, Anthropic
Background: 10 years as an embedded C++ developer. Firmware work teaches you to plan for what breaks at 3am: timeouts, retries, malformed input, flaky networks. Scrapers, bots, and websites fail the same ways, so mine ship with those cases already handled.
Everything ships with source code you own, a short setup doc, and tests. My code is public on GitHub: async parallel scraping with a web UI and image search (price-aggregator), Gmail API + OAuth2 extensions (keeptouch), Manifest V3 service workers with reconnect logic (sse-notifier-extension), n8n + LLM pipelines (upwork-job-alert), plus three hand-coded sites live at 97 to 100 on mobile PageSpeed (salon.helban.dev, kalkulator.helban.dev, dropwatch.helban.dev) and a WordPress speed case study (wp-speed.helban.dev). That price scraper runs 24/7 as a Docker container on my own server. Writing a scraper is one thing, keeping it running is another, and I do both.
I work async: written updates and fast replies, no calls needed. Send a link or a sample file with your message. On most jobs I run a smoke test on your actual data first, so you see real output before committing to full scope.
Steps for completing your project
After purchasing the project, send requirements so Adam can start the project.
Delivery time starts when Adam receives requirements from you.
Adam works on your project following the steps below.
Revisions may occur after the delivery date.
Analyze requirements and design prompt engineering strategy
Develop the Python AI tool or autonomous agent architecture