search

vineetUpadhayay / sunbird-va-bot-service

1 stars 4 forks source link

[DMP 2024]: Integration of retraining pipelines to improve bot responses #1

Open vineetUpadhayay opened 1 month ago

vineetUpadhayay commented 1 month ago

Ticket Contents

Description

The service here is a RAG based virtual assistant which can be fine-tuned for user specific requirements. The bot uses pre-loaded documents like pdf, audio, video etc. which on the basis of similarity search provides specific information along with the source of information. Currently the bot uses these documents and GPT to provide the condensed responses.

For this issue we want to cache the GPT responses along with the documents as short term / long term memory so that we can use these responses to provide more improved and customised responses.

Goals & Mid-Point Milestone

Goals

Setup/Installation

No response

Expected Outcome

No response

Acceptance Criteria

No response

Implementation Details

Mockups/Wireframes

No response

Product Name

Sunbird-Virtual Assistant

Organisation Name

Tekdi

Domain

⁠Service Delivery

Tech Skills Needed

Python

Mentor(s)

https://github.com/vineetUpadhayay

Category

Backend

Harish-2003 commented 4 weeks ago

what technologies do i need to know and are there any beginner friendly issues so that i can able to get familiar with the project code base and stuff thanks @vineetUpadhayay

vineetUpadhayay commented 3 weeks ago

Hi Harish, For starters you need to have basic understanding of python and a python web-framework fastapi

Sanjay8602 commented 3 weeks ago

Hi, I have worked with python before and currently also working as AI intern (also deep diving in it) I had the question about how to start to contribute on this project and what tech stack I will need to solve the issues. I can make sure that I will let it down.

Thanks @vineetUpadhayay I will wait for ur response and excited to contribute to it

MustafaAkolawala commented 3 weeks ago

hi @vineetUpadhayay I'm excited to contribute to your project to add caching and improved response customization features to the Sunbird Virtual Assistant. As a backend developer in my previous internship at Apnabot, I have hands-on experience working with large language models (LLMs), Docker, Kubernetes, and Retrieval Augmented Generation (RAG) systems. I'm also well-versed in technologies like Redis, Marqo DB, and FAISS, which are core to this initiative. I believe my expertise in integrating these tools and architectures can help enhance the virtual assistant's capabilities. I'd welcome the opportunity to discuss how I can support the implementation of the caching layer, retrieval-based search, and overall system improvements. Please let me know the best way for me to connect with your team.

Palakiew commented 3 weeks ago

hi @vineetUpadhayay I am writing to express my sincere interest in securing your mentorship for the forthcoming Python project. With a proven track record of Python proficiency, an unwavering commitment to excellence, and a collaborative demeanor, I am poised to make significant contributions to the project's success.

Virajjai commented 3 weeks ago

Hi, @vineetUpadhayay sir, I hope this message finds you well. I've recently come across your project, and I'm really impressed by the work you've done so far. The project aligns perfectly with my interests, and I'm eager to contribute to its growth and success. I have a solid background in Python, having worked with it for around 1 year on various projects involving data analysis with pandas and NumPy. Additionally, I've familiarized myself with like Docker, Kubernetes which I noticed are utilized in your project. I believe my experience in these areas can be beneficial in contributing effectively to your project. I am eager to learn and adapt, ensuring that my contribution aligns with the project's goals and standards. I admire the work you're doing and am excited about the opportunity to be a part of it. Please let me know if you need any more information from me or if there are any steps I should follow to get started.

poozasingh commented 3 weeks ago

Subject : Expression of Interest: Joining Your Team with Python Skills

Hi there, @vineetUpadhayay

After carefully reviewing the objectives and projects, I am enthusiastic about the opportunity to contribute my skills and collaborate with you as your team.

Here's a bit about myself: I am Pooja Singh, currently working in Verana Networks as Software Developer Intern .

My experience includes working on python projects with machine learning in my internship. I'm eager to contact you, explore and work with you.

Regards, Pooja Singh

love1702kush commented 3 weeks ago

Hello @vineetUpadhayay sir, want to be part of this project with a piece of adequate knowledge of Python and I am eager to learn and proceed further.

AzhagurajaR commented 3 weeks ago

Hello @vineetUpadhayay, Very interesting. I want to be involved in and learn the things.

Regards, Azhaguraja.R

glitcher007 commented 3 weeks ago

Hii @vineetUpadhayay

I've been following the progress of our RAG-based virtual assistant closely, and I'm really excited about the idea of making it even smarter with a memory upgrade.

Here's how I'm going to help:

Making Memory Work: I'll team up with you to figure out how to store all the important stuff the virtual assistant learns from GPT, along with the documents it reads. Think of it like saving helpful notes so the assistant can remember them later.

Keeping It Fresh: We'll come up with a plan to keep the memory up-to-date. That means making sure it knows about new documents or changes in information, so it's always giving the most accurate help.

Helping It Remember: I'll work on teaching the assistant how to use its memory when it's talking to users. It'll be like having a handy file cabinet to quickly find the right answers.

Speeding Things Up: We'll make sure the memory doesn't slow down the assistant. It'll be super quick at finding what it needs, so users get help fast.

I'm really looking forward to diving into this project and making our virtual assistant even more awesome. Let's chat soon to get started!

Sunilstar-V commented 2 weeks ago

Hello @vineetUpadhayay I have recently completed this kind of project in my previous project, we developed a virtual assistant that utilized pre-loaded documents such as PDFs, docs, txt, and csv files to provide specific information to users. We integrated the nlp model and GPT for generating condensed responses, and we also implemented caching mechanisms to store GPT responses along with the documents for short-term and long-term memory. This allowed us to provide more improved and customized responses over time. Here we used FAST API and MongoDB to store the data. Given my experience and knowledge in this domain, I think this is also a similar project that I did. am confident that I can make valuable contributions to your project.

Also, would you assign this to me so that I can contribute?

JyothiPriya5 commented 2 weeks ago

Hi @vineetUpadhayay Please provide the basic elements required to contribute this kind of project as beginner and how can I move on further I will wait for your response

RitamhD commented 2 weeks ago

Hi @vineetUpadhayay I am a 2nd year student with prior knowledge in Python. Also I'm having basic knowledge in FastAPI, but I can contribute & lean for it simultaneously as time is not a issue for me. So I am eager to contribute in this project. Let me know from your side about what topics should I focus on for this project. Thank You.

Nandani-sbnk commented 2 weeks ago

Hello sir, I basic knowledge about python. So ,I would like to ask for your guidence.

onceagainarise commented 2 weeks ago

Hello @vineetUpadhayay,

I am a college sophomore with good knowledge in Python, machine learning, and FastAPI, and I have made many projects related to LLMs and machine learning. I am eager to contribute to this project. As a beginner, I would appreciate your guidance.

Regards, Shubham

Alokkr00 commented 2 weeks ago

Hello @vineetUpadhayay,

I am final year student with good knowledge of Python and have some knowledge about FastAPI. Please guide me for this project.

nitishmalang commented 2 weeks ago

hello @vineetUpadhayay I am pre-final year undergrad student I have decent understanding of python I have worked in gsoc 2023 in python project should I proceed with this project

MihirKohli commented 1 week ago

Hello @vineetUpadhayay first of all thanks for this opportunity, i have worked on similar kind of project i do have all the expertise required for above, submitted my proposal. Looking forward to work with you.

karsrinjay2004 commented 1 week ago

Good morning @vineetUpadhayay I am a 2nd year student Of Sir MVIT Bangalore and have fundamental understandings of Python. Please guide me with the project

AbhimanyuSamagra commented 1 week ago

Do not ask process related questions about how to apply and who to contact in the above ticket. The only questions allowed are about technical aspects of the project itself. If you want help with the process, you can refer instructions listed on Unstop and any further queries can be taken up on our Discord channel titled DMP queries. Here's a Video Tutorial on how to submit a proposal for a project.

ananyaatri196 commented 1 week ago

Respected @vineetUpadhayay

Hope this message finds you well. I am a second year student and have a keen interest into this project. I have been using Python for more than couple years now. Looking forward to working on this project.

Omkarj00 commented 1 week ago

Hey @vineetUpadhayay I'm Omkar, am interested in contributing to project! My skills in machine learning, NLP, and Python are a great fit. I'm eager to explore machine learning pipelines

Is there a way you prefer for me to reach out? I'm looking forward to exploring how I can contribute.

renuka010 commented 1 week ago

Hi @vineetUpadhayay I would like to work on this ticket. I have strong knowledge of Python and FastAPI and also have created RAG applications using vector databases. I would like to optimize the pipeline. I noticed there isn't any dedicated channels for Tekdi in Discord. Is there a way to connect with mentors?

sweta2484 commented 1 week ago

Hi @vineetUpadhayay I am writing to express my keen interest in joining the retraining pipelines integration project. With a strong background in AI and natural language processing, I am eager to contribute to this innovative initiative. My experience in the relevant skills positions me well to make valuable contributions to the project. I am excited about the opportunity to work under your mentorship and contribute to the success of this project. I would greatly appreciate the chance to discuss this opportunity further.

SashithaThummagunta commented 1 week ago

Hi @vineetUpadhayay I have knowledge on python development . can u please guide me how to contribute ?

PriyalPB commented 1 week ago

Hi @vineetUpadhayay ! I'm a third year student from Cummins Pune.

I'm thrilled to join your Integration of retraining pipelines to improve bot responses project and offer my skill sets which has a strong background in Machine Learning ,deep learning (CNN), NLP and Python, which seem to fit perfectly with what you're looking for. I'm excited to explore how my expertise can elevate the project. Furthermore, the integration computer vision along with the ML advancements could lead to a seamlessly automated system. I'm eager to discuss further avenues where I can make meaningful contributions. Could we schedule a meeting to delve into this in more detail?

SashithaThummagunta commented 1 week ago

Sure we can discuss about it .

On Sun, 28 Apr, 2024, 3:04 pm PriyalPB, @.***> wrote:

Hi @vineetUpadhayay https://github.com/vineetUpadhayay ! I'm a third year student from Cummins Pune.

I'm thrilled to join your Integration of retraining pipelines to improve bot responses project and offer my skill sets which has a strong background in Machine Learning ,deep learning (CNN), NLP and Python, which seem to fit perfectly with what you're looking for. I'm excited to explore how my expertise can elevate the project. Furthermore, the integration computer vision along with the ML advancements could lead to a seamlessly automated system. I'm eager to discuss further avenues where I can make meaningful contributions. Could we schedule a meeting to delve into this in more detail?

— Reply to this email directly, view it on GitHub https://github.com/vineetUpadhayay/sunbird-va-bot-service/issues/1#issuecomment-2081410451, or unsubscribe https://github.com/notifications/unsubscribe-auth/A4RPSX4LNW44QQHNQDZFCILY7S7CRAVCNFSM6AAAAABFYQYYOGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOBRGQYTANBVGE . You are receiving this because you commented.Message ID: @.***>

PriyalPB commented 5 days ago

How can I contact you?

On Sun, Apr 28, 2024, 3:14 PM Thummagunta .Sashitha < @.***> wrote:

Sure we can discuss about it .

On Sun, 28 Apr, 2024, 3:04 pm PriyalPB, @.***> wrote:

Hi @vineetUpadhayay https://github.com/vineetUpadhayay ! I'm a third year student from Cummins Pune.

I'm thrilled to join your Integration of retraining pipelines to improve bot responses project and offer my skill sets which has a strong background in Machine Learning ,deep learning (CNN), NLP and Python, which seem to fit perfectly with what you're looking for. I'm excited to explore how my expertise can elevate the project. Furthermore, the integration computer vision along with the ML advancements could lead to a seamlessly automated system. I'm eager to discuss further avenues where I can make meaningful contributions. Could we schedule a meeting to delve into this in more detail?

— Reply to this email directly, view it on GitHub < https://github.com/vineetUpadhayay/sunbird-va-bot-service/issues/1#issuecomment-2081410451>,

or unsubscribe < https://github.com/notifications/unsubscribe-auth/A4RPSX4LNW44QQHNQDZFCILY7S7CRAVCNFSM6AAAAABFYQYYOGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOBRGQYTANBVGE>

. You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/vineetUpadhayay/sunbird-va-bot-service/issues/1#issuecomment-2081413205, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYKRXWMV33IOWFZ4EHS2TYLY7TAHPAVCNFSM6AAAAABFYQYYOGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOBRGQYTGMRQGU . You are receiving this because you commented.Message ID: @.***>

chakrapriya7 commented 5 days ago

Hello @vineetUpadhayay ,Iam a final year student in Computer Science and Engineering.I have been experiencing python since 4years and Iam looking forward to contribute this project as my interest lies .Thankyou.

Nivedita-MN18 commented 5 days ago

I recently came across the Sunbird-Virtual Assistant project and was intrigued by its aim to develop a RAG-based virtual assistant with enhanced caching capabilities for improved and customized responses.

As someone with a strong background in Python and an interest in backend development, I am excited about the opportunity to contribute to this project. The goals outlined, particularly in setting up a caching layer, implementing retrieval from cached documents using similarity search, and dockerizing the service, align closely with my skills and interests.

I have experience working with technologies such as Redis for caching, Marqo DB for vector storage, and Docker for containerization. Additionally, I am eager to learn more about FAISS-based similarity search, microservices architecture, and Kubernetes deployment.

Could you please provide guidance on how I can get involved in the project? Whether it's contributing to setting up the caching layer, implementing retrieval algorithms, or assisting with Dockerization and deployment, I am eager to contribute in any way I can.

Thank you for considering my interest in joining the Sunbird-Virtual Assistant project. I look forward to the possibility of working together to enhance the capabilities of this innovative virtual assistant.

ItshMoh commented 5 days ago

Hey @vineetUpadhayay I want a clarification regarding the sessions and caching. Will the user have multiple sessions, in each session he has some chat with the document or other sources. Sessions can have the same source of information(Documents) or different. He would be able to access his sessions and can continue the chat.

Mohan :)

Divyashetty333 commented 5 days ago

Hi @vineetUpadhayay I am 3rd year student in GNITS and working with python and interested to work on this ticket

skddl007 commented 4 days ago

In a previous project, I led the development of a virtual assistant leveraging various document formats like PDFs, docs, txt, and csv files to offer tailored information to users. We integrated NLP models and GPT for concise responses, while implementing caching mechanisms to enhance response quality over time. Utilizing FAST API and Postgresql for flexible data management, I believe my expertise aligns well with the current project, allowing me to offer significant contributions.

worrier1728 commented 3 days ago

Hello maam, I hope this message finds you well. I am Isha and I had worked as an AI Trainer and Data Engineer at Soul AI. According to me the solution for this problem to me is: "Implementing a caching system (short-term and long-term) to store GPT responses alongside documents. This will enable the bot to leverage past responses for improved customization." Could you please provide guidance on how I can get involved in the project? Whether it's contributing to setting up the caching layer, implementing retrieval algorithms, or assisting with Dockerization and deployment, I am eager to contribute in any way I can. Thank you for considering my interest in joining the Sunbird-Virtual Assistant project. I look forward to the possibility of working together to enhance the capabilities of this innovative virtual assistant.