You will get Computer Vision Proof of Concept
Top Rated
Top Rated
Project details
You will get an MVP for your computer vision idea. With over 7 years of experience from some of the biggest AI labs in the world, I will assure you, you will get the bang for your buck!
What's included
| Service Tiers |
Starter
$499
|
Standard
$999
|
Advanced
$1,499
|
|---|---|---|---|
| Delivery Time | 7 days | 14 days | 21 days |
Number of Revisions | 0 | 2 | 5 |
Number of Model Variations | 2 | 3 | 4 |
Model Validation/Testing | - | ||
Model Documentation | - | - | |
Data Source Connectivity | - | ||
Source Code |
Optional add-ons
You can add these on the next page.
Fast Delivery
+$499 - $999Frequently asked questions
11 reviews
(11)
(0)
(0)
(0)
(0)
This project doesn't have any reviews.
SH
Spencer H.
Feb 24, 2026
Expert Machine Learning Engineer for LLM Training
SP
Sebastian P.
Jan 11, 2026
Computer Vision Architect for DeepStream on Jetson
Working with Mayur Kulkarni and Adagrad AI has been one of our most successful professional engagements. We initially hired Mayur to consult on our NVIDIA DeepStream environment on Jetson, but his role quickly evolved as we recognized his immense technical depth.
Most recently, he took on the complex task of developing a Triton Inference Server from the ground up to handle our proprietary marker recognition. This required him to integrate his work seamlessly with a DeepStream pipeline managed by a separate team—a task he handled with precision and clear communication.
If you are looking for someone who understands the intersection of high-level AI research and hardware-specific deployment (specifically NVIDIA’s stack), Mayur is the gold standard. We highly recommend him for any mission-critical Computer Vision or DeepStream architecture work.
Most recently, he took on the complex task of developing a Triton Inference Server from the ground up to handle our proprietary marker recognition. This required him to integrate his work seamlessly with a DeepStream pipeline managed by a separate team—a task he handled with precision and clear communication.
If you are looking for someone who understands the intersection of high-level AI research and hardware-specific deployment (specifically NVIDIA’s stack), Mayur is the gold standard. We highly recommend him for any mission-critical Computer Vision or DeepStream architecture work.
JS
Josh S.
Jun 23, 2025
Voice Agent Pipeline for B2B app
Mauyur was great on our project and very helpful! I would use him again in the future!
AH
Almas H.
Apr 1, 2025
Seeking Data Scientists with ETL, ML, Python experience / gsmlsmus-2025-q1
AA
Antonio A.
Mar 24, 2025
1hr Video Tutoring Help setting up dual LLMs on a small 48GByte VRAM Server using vLLM
Super friendly, super knowledgeable. Would hire again.
About Mayur
GenAI | Computer Vision Expert | Ex-Microsoft Research
98%
Job Success
Oslo, Norway - 11:16 am local time
I'm an experienced Computer Vision Researcher, and LLM researcher and have experience in working at some of the best AI labs in the world. In my previous job at Microsoft Research, I put large scale Computer Vision models into production in the Azure ML team, and also designed highly scalable APIs in Python, C, and C++. My models are deployed at locations with 10,000+ sites worldwide. Before that, I worked on automated video analysis to extract key moments from large number of videos at IIT Bombay, one of the world's leading Computer Vision lab. I have also been featured by NVIDIA at their annual GTC 2021 Asia, and GTC 2022 Silicon Valley
I have over 10 years of hard core research experience in Computer Vision, Deep Learning, Reinforcement Learning, and Natural Language Processing. I have authored a book in Applied Reinforcement Learning that goes by the title "Reinforcement Learning Workshop" by Packt Publications. Furthermore, I have also contributed in very well known open source Python libraries like scikit-learn, DVC, and PyTorch.
Key strengths and skills:
🦾 Extensive (50+ projects) Prompt Engineering experience with many different LLMs and systems like OpenAI GPT-4o/o1, Claude 3.5 Sonnet/Opus, Mistral, Gemini, Llama 3, HuggingFace, Together.ai, Ollama, LM Studio, Groq and many others
🦾 Skilled in getting AI projects to production with frameworks like TextGrad, DSPy, ell, Langchain, LlamaIndex, LangServe, LangSmith, Kor, Zapier, MemGPT, Claude Computer Use and other cutting-edge AI tools
🦾 Achieved dramatic cost reduction and runtime improvements with fine tuning using QLoRA, DeepSpeed, OpenAI GPT-3.5, Unsloth, Quantization, Transformers and GPU clusters
🦾 Successfully completed multiple Agentic framework projects with Claude Sonnet 3.5 Computer Use, OpenAI Swarm, Autogen, CrewAI, AutoGPT, LangGraph, MultiOn, Claude Computer Use, Open Interpreter and other agent-based systems
🦾 Strong background in computer vision and generative imaging/video with OpenCV, Stable Diffusion, Pika, Runway, Opus and Midjourney.
🦾 Built real time voice speech to text and text to speech systems with Twilio, Bland.ai, Whisper, ElevenLabs and others
🦾 Proficient in ML frameworks (TensorFlow, Keras, PyTorch, scikit-learn) and programming languages (Python, Mojo, C++, Javascript)
🦾 Expertise in algorithmic trading, stock/option/future forecasting, cryptocurrency, NFTs, DeFi protocols and Solidity with time series technology like AutoTS, Prophet, TimeGPT, PatchTST, etc.
🦾 Skilled in full-stack web development (V0.dev, React, Angular, Vue.js, Django, Node.js, FastAPI, Flask) and cloud computing (AWS, GCP, Azure, DigitalOcean)
🦾 Proficient in MLOps tools (Weights and Biases, Neptune, MLFlow, BentoML, Docker) and databases (RAG, Faiss, Weaviate, Pinecone, Neo4J, MongoDB, MySQL, BigQuery, PostgreSQL, SQLServer)
Are you ready to apply high-quality AI in your company? If you're looking for a quality researcher, feel free to drop a line!
Steps for completing your project
After purchasing the project, send requirements so Mayur can start the project.
Delivery time starts when Mayur receives requirements from you.
Mayur works on your project following the steps below.
Revisions may occur after the delivery date.
Receive scope of work from the client
Here, you'd tell me what's your use case and how do you want me to deliver your proof-of-concept
Get data (OR we gather it for you)
Once we've got down the scope, now you give us your data (which might be already labeled), OR you can get it labeled using my team. Using our service, I can get you your first 500 images labeled free of cost.