You will get a custom computer vision model and deploy it as an API
Rising Talent
Project details
Off-the-shelf models rarely fit real-world problems. I build computer vision systems trained specifically on your data — then deploy them so they're actually usable.
With 6+ years in Python and AI/ML, I've built and shipped vision systems including a real-time YOLOv8 surveillance detector and a medical X-ray classifier achieving 96%+ accuracy. I know what it takes to go from raw images to a reliable, fast model in production.
What sets this project apart:
→ Custom-trained model on your dataset — not a generic pretrained demo
→ Full pipeline: data prep, training, evaluation, and optimization
→ Deployed as a clean FastAPI endpoint your app can call instantly
→ Works with images, video files, or live camera streams
→ Detailed evaluation report with accuracy metrics included
Whether you need to detect objects, classify images, or analyze video in real time, I'll deliver a model that performs — and an API that's ready to use on day one.
With 6+ years in Python and AI/ML, I've built and shipped vision systems including a real-time YOLOv8 surveillance detector and a medical X-ray classifier achieving 96%+ accuracy. I know what it takes to go from raw images to a reliable, fast model in production.
What sets this project apart:
→ Custom-trained model on your dataset — not a generic pretrained demo
→ Full pipeline: data prep, training, evaluation, and optimization
→ Deployed as a clean FastAPI endpoint your app can call instantly
→ Works with images, video files, or live camera streams
→ Detailed evaluation report with accuracy metrics included
Whether you need to detect objects, classify images, or analyze video in real time, I'll deliver a model that performs — and an API that's ready to use on day one.
AI Development Type
Deep Learning, Model Tuning, Software MaintenanceAI Tools
Amazon SageMaker, Azure Machine Learning, BigDL, Google AutoML, Keras, NVIDIA AI Platform, OpenCV, PyTorch, TensorFlowAI Development Language
PythonWhat's included
| Service Tiers |
Starter
$300
|
Standard
$700
|
Advanced
$1,500
|
|---|---|---|---|
| Delivery Time | 5 days | 14 days | 21 days |
Number of Revisions | 1 | 2 | 3 |
AI Model Integration | |||
Detailed Code Comments | - | ||
Knowledge Graph | - | ||
Model Documentation | - | ||
Ontology | - | - | |
Source Code | - | - | |
Taxonomy | - | - |
4 reviews
(4)
(0)
(0)
(0)
(0)
This project doesn't have any reviews.
KA
Khulood A.
Dec 9, 2024
Senior Django developer
He is very professional and talented, I'm planning to stick to him to work together on any other projects.
KH
Karim H.
Oct 26, 2024
Need senior python developer
Otabek exceeded all expectations on our recent project. They demonstrated an impressive command of Python, tackling complex challenges with efficiency and precision. Their code was clean, well-documented, and optimized, which significantly improved our project’s performance. Otabek also exhibited excellent problem-solving skills and communicated effectively, keeping everyone informed at every stage. Their proactive approach and attention to detail made them a valuable asset to our team. We look forward to collaborating with them again!
SC
Sean C.
Sep 30, 2024
AI-Powered Essay Evaluation / Assessment
Work completed successfully. Would work with again
MG
Michail G.
Sep 13, 2024
Solving a python script issue in DWSIM
Quick response, easy communication, will hire again
About Otabek
ML Engineer | Voice AI & LLM & RAG Deployment | Python, FastAPI, CUDA
100%
Job Success
Namangan, Uzbekistan - 11:48 pm local time
Most of my work is one of three things:
Voice agents that hold a conversation. I spent 16 months building a voice-interactive companion app for dementia care — AI-generated personas, full STT/TTS pipeline, and conversational memory drawn from each patient's background, likes and life events. Latency and turn-taking are what make a voice agent feel human or feel broken, and that is the part I engineer rather than configure.
Private and on-prem LLM deployment. I have put a self-hosted LLM stack onto a client's own Ubuntu server with an NVIDIA GPU — resolving CUDA and dependency conflicts, then handing over installation and implementation documentation so their team could run it without me. If your data cannot leave your network for legal or policy reasons, this is the work.
LLM pipelines and evaluation at volume. I built a system that scored 7,000 academic essays in a single week — ingesting PDF and DOC files from cloud storage, running GPT-4 against a rubric-derived prompt tuned to the client's tone of voice, batch-processing with logging and error handling, spot-check QA, and emitting per-essay scores, written feedback and rankings. Delivered a month ahead of deadline.
What clients have said:
"Otabek exceeded all expectations. They demonstrated an impressive command of Python, tackling complex challenges with efficiency and precision. Their code was clean, well-documented, and optimized, which significantly improved our project's performance."
"He is very professional and talented, I'm planning to stick to him to work together on any other projects."
Core stack: Python, FastAPI, PyTorch, CUDA, OpenAI API, RAG and vector databases, Django, PostgreSQL, Docker.
I hold a 100% Job Success Score, I reply within a few hours, and I will tell you early and plainly when something in a spec won't work — before it costs you a sprint.
Send me your project details and I'll tell you honestly whether it's a fit.
Steps for completing your project
After purchasing the project, send requirements so Otabek can start the project.
Delivery time starts when Otabek receives requirements from you.
Otabek works on your project following the steps below.
Revisions may occur after the delivery date.
Data review & planning
Assess your dataset, define classes, and select the best model architecture for your use case.
Data preparation
Clean, augment, and split the dataset into train/validation/test sets.