Upwork home
Sign up
Project Catalog
DesignDesign
Video & AudioVideo & Audio
Development & ITDevelopment & IT
MarketingMarketing
Writing & TranslationWriting & Translation
MoreMore

You will get recommendation system/semantice search engine with text or image embeddings

Spike X.Status: Offline
Spike X.
Preview image

Let a pro handle the details

Buy Machine Learning services from Spike, priced and ready to go.
Spike X.Status: Offline
Spike X.
Preview image

Let a pro handle the details

Buy Machine Learning services from Spike, priced and ready to go.

Project details

I have developed a recommendation system based on over 100 million data of semantic search with Faiss, and optimized word embedding computation with Milvus.
I am happy to choose the best practice for the project no matter how complicated the solution is.
Machine Learning Tools
BERT, MLflow, NumPy, Python, PyTorch, SQL
What's included
Service Tiers Starter
$3,500
Standard
$4,000
Advanced
$4,500
Delivery Time 30 days 35 days 40 days
Number of Revisions
000
Model Validation/Testing
Model Documentation
Data Source Connectivity
Source Code
Spike X.Status: Offline

About Spike

Spike X.Status: Offline
Data Science/ML Engineer
Shenzhen, China - 1:28 pm local time
I’m an experienced ML engineer/data scientist/software engineer, problem solver, and good chat partner. Expertise in developing and deploying machine learning models and systems. Skilled in Python, deep learning frameworks, and SOTA models. 6+ years of experience in developing, machine learning, and data engineering.

Skills
Languages: Python, JavaScript, Dart
Frameworks and packages: FastAPI, Pytorch, Transformers, RASA, Detectron2, Spacy
Tools: Docker, Git, Supervisor, CI/CD, MLflow
Database: Mysql, Redis, Elasticsearch, Milvus, MongoDB

Data&Engineering
● Build backend services with FastAPI
● Designed a distributed crawl system with Scrapy, customized middle pipelines to format the data.
● Construct a machine learning architecture with MLflow, FastAPI, CI/CD
● Synchronized data from Mysql to ElasticSearch by Logstash, and designed the index structure of ES

NLP
● Developed a recommendation system based on over 100 million semantic search data with Milvus.
● Finetune Mistral 7B with PEFT in Lora method to develop an AI assistant.
● Classified texts including policies, news, and recruitment messages with Bert-like models.
● Built a task-oriented chatbot with Rasa framework, classified different intends with NLU modules in Rasa, and deployed the robot with docker-compose.

CV
● Developed object detection model for an image check website
● build various functions with OpenCV

Steps for completing your project

After purchasing the project, send requirements so Spike can start the project.

Delivery time starts when Spike receives requirements from you.

Spike works on your project following the steps below.

Revisions may occur after the delivery date.

the API of vector database

offer the API for vector database

document

document about the API and database

Review the work, release payment, and leave feedback to Spike.