You will get a fully trained deeplearning/ML Model
Rising Talent
Project details
You will get a fully trained ML / DL model with an interface to run the model and get predictions. I will develop end to end deep learning and machine learning systems for Classification, Clustering, Correlation, and Regression Analysis, ComputerVision and NLP
I got experiences and expertise in the following domains,
1)Machine learning classification
2)Regression
3)Natural Language processing
4)Deep learning
5)Computer vision
6)SVM
I am capable of using
Tools and Technologies:
1)Python
2)OpenCV
3)Tensorflow
4)Keras
5)Flask => as per requirements
6)React / Html => as per requirements
7)pytorch => as per requirements
8)pandas
9)Numpy
12)Scikit-Learn
14)Matplotlib
Things that are important to me
1) Prioritize Client Satisfaction
2) Future Support on Completed Orders
3) Unlimited Revisions
4) 24/7 available
I appreciate your interest, and thanks for reading my profile!
If you get a chance, come visit to discuss your next project.
I would love to help you with your NLP and Computer Vision Models.
I got experiences and expertise in the following domains,
1)Machine learning classification
2)Regression
3)Natural Language processing
4)Deep learning
5)Computer vision
6)SVM
I am capable of using
Tools and Technologies:
1)Python
2)OpenCV
3)Tensorflow
4)Keras
5)Flask => as per requirements
6)React / Html => as per requirements
7)pytorch => as per requirements
8)pandas
9)Numpy
12)Scikit-Learn
14)Matplotlib
Things that are important to me
1) Prioritize Client Satisfaction
2) Future Support on Completed Orders
3) Unlimited Revisions
4) 24/7 available
I appreciate your interest, and thanks for reading my profile!
If you get a chance, come visit to discuss your next project.
I would love to help you with your NLP and Computer Vision Models.
What's included
| Service Tiers |
Starter
$50
|
Standard
$80
|
Advanced
$120
|
|---|---|---|---|
| Delivery Time | 5 days | 7 days | 12 days |
Number of Revisions | Unlimited | Unlimited | Unlimited |
Number of Model Variations | 1 | 2 | 3 |
Number of Graphs/Charts | 2 | 4 | 5 |
Model Validation/Testing | |||
Model Documentation | |||
Data Source Connectivity | - | - | - |
Source Code |
29 reviews
(29)
(0)
(0)
(0)
(0)
RA
R A.
Dec 13, 2021
Roshan is very good at understanding the requirement and provide the quality work. I recommend him to sovle problems related to Ml OR DL."
VV
Vishnu V.
Apr 15, 2024
DL for genomics
Very positive.
PR
Prudhvi R.
Sep 29, 2023
Python tasks
Good python skills.Very quick. HIGHLY RECOMMENDED!!
PR
Prudhvi R.
Sep 22, 2023
Excel + Python
Outstanding freelancer. Very well versed in Python, Machine learning.
NY
Nour Y.
Sep 22, 2023
Website that integrates ChatGpt Api and Generates files
AG
Adrian G.
Apr 17, 2023
Scrap and store a website data
Roshan is one of the best guy to work with. He is very patience to hear our requirements and executed it swiftly. We will definitely engage him again. Thank you!
About Roshan
Generative AI Engineer | RAG, LLM Agents, Python, AWS
Islamabad, Pakistan - 2:10 am local time
Backed by 40 completed Upwork jobs and 5+ years of AI/ML and software engineering experience, I help startups, SaaS companies, and product teams build reliable RAG applications, AI agents, LLM-powered APIs, and scalable Python services on AWS.
Whether you need to build an AI product from scratch, productionize an existing prototype, improve the accuracy of a RAG system, reduce LLM costs and latency, or deploy models securely, I can help.
I work across the complete development lifecycle:
Technical discovery → system architecture → development → testing → deployment → monitoring → optimization
𝗪𝗛𝗔𝗧 𝗜 𝗖𝗔𝗡 𝗛𝗘𝗟𝗣 𝗬𝗢𝗨 𝗕𝗨𝗜𝗟𝗗
• RAG and enterprise knowledge assistants
Document ingestion, chunking, embeddings, vector search, retrieval optimization, context ranking, citations, and grounded LLM responses.
• AI agents and workflow automation
Tool calling, multi-step reasoning, API integrations, structured outputs, validation, human approval workflows, and automated task execution.
• LLM and Generative AI integrations
OpenAI, Hugging Face Transformers, LangChain, LlamaIndex, PydanticAI, Ollama, llama.cpp, and MCP-based tool integrations.
• Secure Python backends and APIs
FastAPI, Django, Flask, REST APIs, background tasks, authentication, PostgreSQL, MySQL, MongoDB, and third-party integrations.
• AWS AI and ML infrastructure
Amazon Bedrock, SageMaker, Lambda, ECS, EKS, S3, RDS, OpenSearch, and scalable real-time inference architecture.
• Private and local AI deployments
Quantized LLMs, CPU-based inference, Dockerized model serving, privacy-focused deployments, and reduced cloud dependency.
• Production deployment and DevOps
Docker, Kubernetes, Terraform, GitHub Actions, CI/CD, Prometheus, Grafana, CloudWatch, automated testing, and infrastructure as code.
𝗔𝗜 𝗥𝗘𝗟𝗜𝗔𝗕𝗜𝗟𝗜𝗧𝗬 𝗔𝗡𝗗 𝗢𝗣𝗧𝗜𝗠𝗜𝗭𝗔𝗧𝗜𝗢𝗡
A working AI demo is only the beginning. I can also help with:
• Structured and schema-validated LLM outputs
• Hallucination reduction and retrieval-quality improvement
• Guardrails, error handling, and fallback workflows
• Prompt and context optimization
• Model selection and evaluation
• LLM cost and latency optimization
• Logging, tracing, monitoring, and observability
• Scalable and maintainable system architecture
𝗧𝗬𝗣𝗜𝗖𝗔𝗟 𝗘𝗡𝗚𝗔𝗚𝗘𝗠𝗘𝗡𝗧𝗦
• Building an AI MVP from an idea or specification
• Converting an AI prototype into a production-ready application
• Adding AI capabilities to an existing SaaS or web product
• Auditing and improving an existing RAG or agent system
• Developing a FastAPI backend for an AI application
• Deploying AI services securely on AWS
• Troubleshooting inaccurate, slow, or expensive LLM workflows
𝗪𝗛𝗔𝗧 𝗬𝗢𝗨 𝗥𝗘𝗖𝗘𝗜𝗩𝗘
Depending on the project, deliverables may include:
• System architecture and implementation plan
• Clean, maintainable source code
• Secure APIs and database integrations
• Automated tests and validation
• Docker and deployment configuration
• CI/CD pipelines and infrastructure code
• Monitoring and error-handling setup
• Technical documentation and knowledge transfer
𝗥𝗘𝗖𝗘𝗡𝗧 𝗘𝗫𝗣𝗘𝗥𝗜𝗘𝗡𝗖𝗘
My recent work includes production RAG pipelines, tool-using agentic workflows, schema-validated AI services, privacy-focused local LLM inference, AWS-hosted ML infrastructure, scalable inference APIs, and full-stack applications powered by Python and React.
𝗪𝗛𝗬 𝗖𝗟𝗜𝗘𝗡𝗧𝗦 𝗪𝗢𝗥𝗞 𝗪𝗜𝗧𝗛 𝗠𝗘
• I design the complete system not just the prompt
• I understand AI, backend engineering, cloud infrastructure, and deployment
• I communicate technical risks, trade-offs, and progress clearly
• I write clean, documented, and maintainable code
• I build solutions that can be monitored, improved, and maintained after launch
• I focus on business outcomes instead of unnecessary technical complexity
Send me a brief description of your goal, current technology stack, available data, and biggest technical challenge. I will help you identify the most practical path from idea to production.
Steps for completing your project
After purchasing the project, send requirements so Roshan can start the project.
Delivery time starts when Roshan receives requirements from you.
Roshan works on your project following the steps below.
Revisions may occur after the delivery date.
Requirement gathering
I will gather all the required requirements for the task e.g data sets
Discussion
We will discuss what approach should be used and will discuss the project in depth