The best works of freelancers in machine learning on Freelancehunt — examples of work performed in machine learning

Switch to English version?

Yes

Переключитись на українську версію?

Так

Переключиться на русскую версию?

Да

Przełączyć się na polską wersję?

Tak

Login
Registration
- Welcome to Freelancehunt
  
  Work risk-free, saving time and money
  
  Login Registration

NaftoGaz Europe Chateau News Site

AI & Machine Learning

by php #yii2

Peiko17
22 USD
Recognition of cats and dogs on images

AI & Machine Learning

A demonstration program to recognize cats and dogs on the images. Strong neuronal networks. The accuracy is 85-90%. The interface is made on Tkinter.

slado122
AI (Ai)

AI & Machine Learning

This image was created at my request in Midjourney for the client.

stayshadesigner
223 USD
AI Specialist Certificate

AI & Machine Learning

Completed a full course on AI.

anna_liv
33 USD
Funny news

AI & Machine Learning

Neural network generating funny news based on real news of Ukraine from https://korrespondent.net/ukraine/
To generate news, write /newh bot

Jobgter
223 USD
Automatic Removal of WaterMarks Logos

AI & Machine Learning

Based on machine learning methods, models for automatic search and removal of the logo (WaterMark, waterprint) are developed.The program consists of 3 parts:
Part 1 .Detection, determination of location and cutting the logo with subpixel accuracy.Part 2 .Based on 50-100+ copies of the photo (learning objects) is calculated the reverse mathematical conversion for each pixel, in which the frequency spectrum of the area under the logo corresponds to the area around the logo.This reduces the visibility of the watermark by 90-99% on the original photo.
Part 3 .Package application of the trained watermark removal model for photographs.The Results:
In this work, the logo was removed on 7433 photos.Time of conversion: 3-5 seconds for the cleaning of one photo (1600x1200, 1280x960).The average quadratic deviation is 0.3 gradients of color.Visual control > 93% of photos to recognize the presence of the logo is not possible.6% photo - only with a 300% contrast increase the logo is slightly visible
… 1% photo - when increased on the site of the remote watermark, a small pixel noise is noticed.Price 200-300$ - training model 1 type of logos
(depending on the complexity of the logo)
N x 0.02$ - Automatic processing of 1st photo

(Examples will be submitted on request)

webitalex
Assessment of gender pollution

AI & Machine Learning

Console server application that is video (or c live camera or video file)
assess the level of pollution in the selected area.
The whole "must" is classified into 4 categories: dust, small waste, large waste, external waste. All data is recorded in the BD.
Python 3.7 and OpenCV 4

webitalex
3000 USD
AI agent: generation and auto-publishing on WordPress

AI & Machine Learning

# AI Fired the Manager: Generation and Auto-Publication on WordPress

I had a manual process for creating WordPress / Elementor sites: specifications in correspondence, access, templates, texts, images, articles, policies, edits, bug checks, and publication.

… At some point, it became clear: the manager in this scheme becomes a bottleneck. They might forget a detail, pass the task in pieces, confuse priorities, send incomplete specifications, or create urgency where the system is not yet ready.

I decided to eliminate dependence on manual managerial chaos and began to translate the process into an AI workflow.

The scheme became:

**Specifications / config → Gemini → texts → images → QC → pass2 → cleanup → WordPress / Elementor → publication**

Instead of manually assembling the site from scratch each time, the system takes structured data, generates content through Gemini, checks the result, cleans up AI junk, and prepares for publication on WordPress.

Two models are used internally:

* `gemini-2.5-flash` — texts, articles, processing problematic fragments;
* `gemini-2.5-flash-image` — images for the site.

Python manages the process as an orchestrator: it launches stages, saves data, checks results, sends problematic areas for a second AI pass, publishes content, and writes logs.

The most important part is quality control. AI can generate text, but it might leave a placeholder, lorem, an old phrase from the template, extra characters, or an almost unchanged block. Therefore, the project has a QC chain:

**suspicious detect → pass2 through Gemini → lorem cleanup → vacuum-cleanup**

Protective elements have been added separately: `STOP_NOW.txt`, `protected_domains.txt`, retry, handling 429, `.env`, `GEMINI_API_KEY`, `project_config.json`.

Result: AI began to perform the part of the work where managerial control, manual copy-pasting, and constant clarifications were previously needed.

This is not just a prompt for generating text. This is a working pipeline:

**LLM → Python orchestration → QC → WordPress / Elementor → auto publish**

Manual work on the site has been broken down into stages, measured, and partially automated. The process has begun to move from "the manager keeps everything in their head" to a system where data, generation, checking, and publication follow a clear chain.
#AIagent #AIworkflow #Gemini #Python #WordPress #Elementor #Automation #AIautomation #AIautomation #LLM #GoogleGemini #ContentAutomation #AutoPublication #WordPressAutomation #PythonAutomation

Me4Hik
1200 USD
AI Assistant for Call Automation in a Dental Clinic

AI & Machine Learning

An automation system was implemented to process incoming and outgoing calls for a dental clinic. The solution not only records every call but also converts it into structured data for further work with patients and quality control of administrators. The system receives call data from Binotel, performs speech-to-text transcription, analyzes conversations, sends results to Cliniccards CRM, and generates weekly analytics for all interactions.

How the solution works:

… 1) The system automatically receives data for every incoming and outgoing call from Binotel.
2) Call recordings are automatically converted into text for further analysis.
3) Based on the conversation, the system identifies the main intent:
appointment booking, consultation, rescheduling, interest in specific services, and other details.
4) Data is automatically transferred to Cliniccards CRM:
- If a patient does not exist — a new record is created.
- If the patient already exists — their data is updated.
5) The system evaluates call quality based on predefined criteria:
greeting, clinic introduction, правильні питання, correct closing of the conversation.
6) Weekly reports are generated with key metrics:
number of new patients, most common requests, communication quality, and other KPIs.

Key business benefits:

- Improved call handling quality
No calls are missed or left unanalyzed. Management gains full visibility of staff performance.

- Automatic CRM data population
All call data is automatically saved in patient profiles without manual input.

- Staff performance control
The system evaluates communication quality and helps identify improvement areas.

- Deep call analytics
The clinic receives insights into patient needs, popular services, and new client flow.

- Time savings
No need to manually listen to calls or update CRM — everything is automated.

Conclusion:
Call automation enabled the clinic to combine telephony, AI analysis, and CRM into a single system. The business now has full control over communication, structured data, and actionable analytics for decision-making.

#AI #AIAgent #AIAssistant #OpenAI #GPT #n8n #Makecom #Automation #BusinessAutomation #Binotel #Ringostat #Twillio #CRM #CallAnalysis #SpeechToText #Transcription #DentalClinic #Automation #calltracking

d4rk51d3n
Interior visualization using AI

AI & Machine Learning

Created a series of realistic images of the interior of a private house using generative artificial intelligence tools. Combined modern style, warm color palette, and natural lighting to create a cozy atmosphere.

What has been done:

… Selected the concept and style of the space.

Generated photorealistic interiors (kitchen, living room, dining room).

Processed details and color accents.

Prepared ready materials for presentations and advertising.

Result for the client:

Unique visualizations for the website, social media, or real estate advertising.
The ability to see the future interior in photorealistic form.

korol_alla
223 USD
Automation | Auto-posting to various social networks

AI & Machine Learning

Task: Creation of a script that will automatically publish posts from a Telegram channel to various social networks (LinkedIn, Facebook, Instagram, Telegram chat, Telegram Stories, Facebook Stories) with content adaptation to the format of each platform and translation into the corresponding language.

Implementation:
1) A Telegram bot has been developed that monitors new posts in the Telegram channel.
… 2) The Telegram bot then sends the text from the posts to Chat GPT, which adapts the content to the format of each platform and translates it into the corresponding language.
3) The final step is auto-posting to social networks.
The system automatically creates a new post in each of the social networks, sends the image from the post in the Telegram channel, and adds the adapted and translated text from Chat GPT.

Development with testing took 2 days.
The client was satisfied with the automation and the time gained for other tasks)

d4rk51d3n
Data Science

AI & Machine Learning

- TensorFlow
- PyTorch
- Pandas
- OpenCV
…
Utilization and finetuning of pre-trained ML models
Consulting on ML/CV integration into your product

shpy

AI & Machine Learning