Sfoglia categorie
Esplora
Fiverr Pro
Italiano
$
USD
I will do machine learning, and ai agents projects using langchain
Digital products developer
Hey! I am Software engineer with 3 production internships. I shipped a feature used by 500+ active users and 10,000+ entries in production. Top 10 in AI Healthcare Hackathon 2026 (100+ teams) with a 1...
Informazioni su questo servizio
Looking for a professional Data Scientist or Machine Learning Engineer? I build custom AI solutions that yield results. With a background in Computer Science and Full-Stack Development, I provide end-to-end integration from training models to building functional web interfaces.
My Services:
- Machine Learning: Regression, Classification, Clustering (Scikit-learn, XGBoost, SVM).
- Deep Learning: CNNs, RNNs, LSTMs (TensorFlow, PyTorch, Keras).
- Data Analytics: EDA, Cleaning, & Visualization.
- NLP & CV: Sentiment Analysis, LLMs (GPT, LangChain), Object Detection (YOLO), & OCR.
- AI Deployment: Building REST APIs (FastAPI/Flask) and embedding models into web/mobile apps.
Tools & Stack:
- Languages: Python, SQL, TypeScript/JavaScript.
- Platforms: Jupyter, Docker, AWS, Google Colab.
- Databases: PostgreSQL, MongoDB, Pinecone (Vector DB).
Why Hire Me?
- Full-Stack Capability: I build both the AI model and the interface to use it.
- Clean Code: Modular, well-documented, and production-ready Python scripts.
- Precision: Deep understanding of the mathematics behind the algorithms.
- Adaptibility: Tailored solutions for your specific needs
Please contact me before ordering to discuss your specific requirements!
Linguaggio di programmazione:
Python
•
R
•
SQL
Framework:
Scikit-learn
•
SimpleCV
•
keras
•
PyTorch
•
Panda
Strumenti:
Quaderno jupyter
•
opencv
•
tensorflow
•
Excel
•
Colab
Il mio portfolio
Altri servizi della categoria Data science e ML offerti da me
FAQ
What types of images can you classify?
Pretty much any domain — medical images (skin lesions, X-rays), product photos, wildlife, documents, satellite imagery, and more. If you have a dataset and a labeling scheme, I can build a model for it.
How many classes can your model handle?
I've worked with up to 12-class classification problems in production settings. I can handle binary through multi-class setups (20+ classes) depending on your dataset size and complexity.
What frameworks and architectures do you use?
PyTorch is my primary framework. I work with proven architectures like EfficientNet, ResNet, and ViT, selecting the best fit based on your data size and accuracy requirements.
What do I need to provide?
Your labeled image dataset (organized by class folders or with a CSV label file), the number of target classes, and any specific accuracy or deployment requirements you have in mind.
What if my dataset is small?
Small datasets are handled with transfer learning from pretrained models, combined with augmentation strategies (flips, rotations, color jitter, and advanced techniques like CutMix/MixUp) to maximize generalization.
What will I receive as a deliverable?
A trained model file (.pt / .onnx), a clean inference script, a training report with accuracy/loss curves and per-class metrics, and instructions on how to run predictions on new images.
Do you sign NDAs or keep my data confidential?
Yes. Your dataset and business context stay private. I'm happy to sign an NDA if required before you share any data.
