Bots in production - Rasa Open Source - Rasa Community Forum
Bots in production
You have selected 0 posts.
21.1k views 84 likes 23 links 73 users
6](https://forum.rasa.com/u/dadecoza "dadecoza")
5](https://forum.rasa.com/u/shubham3582 "shubham3582")
4](https://forum.rasa.com/u/pausali.biswas "pausali.biswas")
4](https://forum.rasa.com/u/Juste "Juste")
3](https://forum.rasa.com/u/harshitazilen "harshitazilen")
read 15 min
Top replies
109 / 109
Aug 2024
post by Juste on Aug 17, 2018
Juste Leader
Have you built a chatbot with Rasa that is now running in production? We would love to know more about it! Share with us what have you built, what were the biggest challenges when deploying it in production, what feedback do you get from the users of your bot, and anything else you would like to share.
Your feedback and experience is what helps us improve Rasa Stack and we believe that your projects will inspire the community to build great chatbots!
Me too
21.1k views 84 likes 23 links 73 users
6](https://forum.rasa.com/u/dadecoza "dadecoza")
5](https://forum.rasa.com/u/shubham3582 "shubham3582")
4](https://forum.rasa.com/u/pausali.biswas "pausali.biswas")
4](https://forum.rasa.com/u/Juste "Juste")
3](https://forum.rasa.com/u/harshitazilen "harshitazilen")
read 15 min
Top replies
post by harshitazilen on Aug 23, 2018
We built a chatbot using Rasa. It is in Alpha Testing mode. We are addressing FAQ data of our website. This is my first project in the field of AI,
I had to work hard while training the bot (data entry in story, domain files). The most important thing to take care is Training Data . I believe if training data is good, bot will answer as expected. Rasa NLU has been very helpful due to it’s pipeline feature that provides flexibility to integrate any NLP library.
Thanks.
post by Juste on Aug 23, 2018
Juste Leader
Awesome! Is it currently running in production? I would love to try it out if there is a chance! Also, did you use both Rasa NLU and Rasa Core for this project?
post by harshitazilen on Aug 23, 2018
Thanks. It uses both Rasa NLU (v 0.12.3) + Core (v 0.9.8). It is live on Slack. Due to NDA with client I may not disclose other details, but YES Rasa has reduced efforts by 40% to build the chat bot. We are quite close to meet client’s expectation (working with optimisations).
13 days later
post by raghibomar on Sep 6, 2018
Hello! I am also developing a bot that is primarily FAQ based for my company. I have developed a basic model on my system and it works well. Now the company wants to move to production phase and since this is my first project, i don’t have much idea about production and would like to know a few things: 1- What are the server requirements in terms of RAM, OS and Memory for a bot that has to handle around 100 users at a time when it is live. 2- What about scaling? Will a single instance of my bot be able to handle all the queries or do i have to use things like docker or wsgi(my front end is in Flask).
16 days later
post by careless25 on Sep 22, 2018
We also use RASA NLU and its components and have several (~70) live bots in production for many different clients. We are currently running into scaling issues since the traffic is getting quiet high.
post by vikasramachandran on Sep 24, 2018
I doubt RASA stack is matured enough to build a production ready chat-bot … I tried migrating from our dialogue flow based chat bot to rasa and faced many hurdles … probably may be I am relatively new to RASA … But I feel that RASA need too much training data to at-least recognize the entities properly …The export from dialog-flow was not enough to migrate and need many modifications to at least to set up basic functionalities to work. and migrations is not seamless :-)… Even some times I see the entities in different user context getting mixed up … Eg: in your weatherbot location supplied by one user taking when a second user ask the same question !!!. Un sure if that was a bug with some specific version of rasa
post by souvikg10 on Sep 24, 2018
souvikg10Souvik GhoshRasa Team
It is a difficult matter to compare Rasa with Dialogflow as with Dialogflow you are probably using many pre-built models to serve your specific task. With Rasa, you will definitely need a bit more understanding of basic chatbot concepts and how to train your models which again demystifies how chatbots can work and how you can control the outcome top-down which in some industries is extremely crucial where risk and liability is higher. Meaning we need to properly explain the inner workings of a chatbot and justify it before taking it to production. For some companies, that may not be a requirement.
I think with the Rasa stack, your team needs a bit more maturity than compared to Dialogflow. I have a bot in production of my own in Dialogflow which works absolutely fine given the need, however I cannot justify the same with my current job given the risk and liability is higher and we need to concentrate a lot more on how exactly are we able to control the outcome.
This is purely based on your needs and the outcome you are looking for in the end So the question of production readiness is purely subjective to your organisational strategy.
post by vikasramachandran on Sep 24, 2018
Companies least likely allow sensitive data or financial data to handle with chat bots even it is hosted on primise. The bots which developed for such companies mostly just like a call center executive answering to users. When some thing come related to critical data mostly bot will handover the chat to actual human.These kind of use cases challenges are least. But if there are requirements which need processing of the collected data , invoking actions from rasa to an external application may not be a good option as it will add more point to point connection and failure handling will be difficult when we add more connection .Not sure if rasa provide any other better way to handle such scenarios.In google dialogue flow each response will have all the data values previously collected for that context and session.
post by souvikg10 on Sep 25, 2018
souvikg10Souvik GhoshRasa Team
In our organisation, all data coming from our client is considered sensitive until qualified and qualifying raw text is really difficult. You can’t really tell your clients not to write sensitive information in a free text field such as chatbots. It doesn’t work. Also, there is always a big question about processing of your data. Google do not play well in such areas.
Like i said it needs the team to mature a bit more to understand how Rasa works. We took some time but for us both cases turn by turn data collection and getting to the right response does work with Rasa. We will be in production soon.
In our case, Rasa has provided us with a better transparency on exactly how our models are performing and upto to which point we can be accurate. We get to evaluate our models something i am not able to do with dialogflow or other chatbot cases. It really boils down to how much control you would like on your chatbot stack for your use cases. For some cases I do agree i don’t need to know that much in depth about how the model is performing since the business case is not that sensitive.
p.s - i am in a financial industry
post by MetcalfeTom on Sep 25, 2018
vikasramachandran](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hey Vikas!
Sorry to hear about that. I’d like to hear a little more about your problems so I can forward them to the team. For the entity recognition, did you try any of the other components in your pipeline listed here?
As for the Dialogflow export, what went wrong exactly? We’re always looking for improvements
post by neerajb1 on Sep 27, 2018
@Juste I had created chat bot on local and want to test it on production. Could you please suggest some documentation about to hosting bot on production. I am very new on development and it will be very helpful for me. Thank in advance
19 days later
post by deepsole on Oct 16, 2018
Hello Juste,
I just created the mood bot and everything is working very well. I have also integrate my working bot into telegram and its working, am able to chat with my bot from the telegram app on my phone.…But the problem is that when the user mood is Sad, it doesnt retrieve the image selected by the user rather bot utters “Did that help?”''this only happens in Telegram""
Am using ngrok and port 5004. I think am getting something wrong some, i also tried running the action server along site with ngrok, but my bot wont still retrieve the image.
I tried changing the url in endpoints.yml but still.it wont show .
Please help
1 month later
post by agrima_agarwal on Nov 21, 2018
Hi,
I am stuck at initial level.
When I m running rasa_core.run with credentials.yml I m stuck at rasa_core server is up and running. May I know after that what are steps to follow, how to open our bot at telegram.
post by adityap31 on Nov 21, 2018
Same here, Lets connect and try to solve the problems.
post by adityap31 on Nov 21, 2018
I am actually working on a assigment to develop a bot for FAQ’s. I am stuck in some places where the main communication integration needs to be done.
post by Juste on Nov 22, 2018
Juste Leader
@agrima_agarwal Hey, have you created the bot on telegram and retrieved the credential details following the guide here? https://rasa.com/docs/core/connectors/#id9
@adityap31 Feel free to open a question with the issues you are facing as well so we can help you out
12 days later
post by Meagol on Dec 4, 2018
raghibomar](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
I would like to know the same things. Did you find answers to your questions? (1- What are the server requirements in terms of RAM, OS and Memory for a bot that has to handle around 100 users at a time when it is live. 2- What about scaling? Will a single instance of my bot be able to handle all the queries or do i have to use things like docker or wsgi(my front end is in Flask).)
1 month later
post by mhjb on Jan 7, 2019
souvikg10](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
For model evaluation, we’ve found QBox.ai very helpful
post by sandeepkumarsahoo on Jan 10, 2019
We have started doing a poc on rasa stack in our company .
Our intention is to handle customers through various channels like voice,mail,sms using Rasa as a backend. We found it can solve our problem . So I gave a demo to my manager 2 days back. He is happy with rasa but he asked me the below questions , Can someone help me to find the answers -
- Is it possible to implement sentiment analysis ? If yes then how ?
- How many requests can be handle by Rasa core and NLU at a time ?
- Is it comparable to Google Dialogflow ?
Thanks in advance.
post by raghibomar on Jan 16, 2019
Meagol](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Sorry for the late reply. My bot is live for ore than a month now and is being used by a lot more than 100 users. Here are the answers to your questions:
I opted for ubuntu server with 15GB RAM(though i read that 2GB was sufficient) and memory of 250GB (again taking more than required).
Nopes, I believe the scaling is built-in in RASA. Even my front-end is on FLASK and i am able to handle multiple users along with different sessions for each user without any explicit scaling.
1 month later
post by kwseow on Feb 27, 2019
Juste](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
hi all, I am also having problem connecting telegram to rasa-core. My server is ok connecting with Slack but just not working with telegram. I followed the setting under the Chat and Voice platforms here ( Chat & Voice platforms) nothing seems to work for telegram. Any advice is deeply appreciated.
post by CandaceRLewis on Feb 27, 2019
agrima_agarwal](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
I’m stuck in mantigames . Anyone can help me? Please.
post by PatrickDS on Feb 27, 2019
sandeepkumarsahoo](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
For an answer to 3, I would say the following : Dialogflow requires a minimal amount of training data, but you need to do everything in their UI and it’s not very flexible in terms of how do you manage your bot. It has the advantage that it requires minimal understanding of AI and very little work, but if you need to justify how good your bot actually is, you’re going to just rely on Google’s brand and not much else. Rasa, on the other hand, gives you more power, and with more power comes more responsibility, i.e. the learning curve is steeper but you end up with a much better bot, in my opinion.
27 days later
post by hemamalini on Mar 26, 2019
hemamaliniRasa Certified Developer
I am trying to integrate Rasa chatbot to rest webapp . I am new to RASA. i have built a bot to trigger the rest call that will take some time to complete the response back .How can i handle such cases where response might take some time . There will be more than 20 users initially for the application . So how can i respond back to the user who initiated that rest call to a web application. Rasa bot can talk to nearly 5 different web applications and how can i present the response back
12 days later
post by dadecoza on Apr 8, 2019
Our bot have recently been deployed to production!
We use the docker containers and I find it to be the easiest way to deploy the Rasa stack.
We currently have 20 custom actions integrating to various systems using Oracle DB, MSSQL, LDAP, REST, AMQP (RabbitMQ) …
We use Git for version control and Jenkins with shell scripts to bring down the containers run the train scripts and spin up the bot when new code is pushed.
Our biggest challenge is to gather intents from the users. Users will type in any nonsense . Every morning we pull the logs and check what users entered the previous day and where the bot were unable to help them, based on that we write intents to cater for those requests (even if the bot cant help them we at least point them to someone who can).
If the bot don’t understand the intention after two tries it will notify a human to assist the user.
So far it has been a fun journey and we are getting good feedback.
Some features …
- Log a request in our ITSM tool (Remedy)
- Query your open and closed requests
- Get incident details
- Submit suggestions
- Request temporary admin access to your device
- FAQ for the general questions (setup email, configure printer, setup vpn …)
- Search for contact details in the company address book
- View your organization structure
- View your company profile
- Find the owner of a device based on asset tag
- Put you in contact with a human if the bot can’t help …
post by Juste on Apr 8, 2019
Juste Leader
Hey @dadecoza. Sounds awesome and congrats on deploying your assistant in production! Is it possible to try it out somewhere? Also, can you tell a little bit more about the domain of your bot?
post by dadecoza on Apr 8, 2019
Hi @Juste,
I work in IT service management for a Telco in South Africa.
The bots main function will be to automate common calls logged to our IT Help Desk. Things like unlocking accounts, giving guest access to the wifi, helping with browser settings etc.
Unfortunately it is internal to our company network but will upload a video of the bot in action when I get a chance.
post by dadecoza on Apr 10, 2019
Here is a quick video …
post by Nishet on Apr 10, 2019
hey @dadecoza nice work.can you share this code if you are not able to share can you please share how to make this chatbot ui? Otherwise can you tell how to do this?share link ya any blog ya any suggstion .Thanks in advance.
post by dadecoza on Apr 10, 2019
Hi @Nishet , We use Skype for Business and that is the interface in the video. I have written a Python Skype for Business Client that integrates to the Rasa bot. GitHub - dadecoza/python-ucwa-bot: Skype for Business Response Bot
post by Nishet on Apr 10, 2019
@dadecoza Ohk got it. Thanks for reply.and i have one more question.do you know how to make ui like skypy ? i mean i want to make ui like that without Skype integration.
post by loicgeek on Apr 11, 2019
dadecoza](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hi @dadecoza, i would like to get in to with you please.
9 days later
post by hossa95 on Apr 21, 2019
@Juste We have deployed Enzo and it’s now working on one of the biggest Car Dealers platforms in
Canada as a sales assistance Used Cars Canda
we were using AWS LEX and now we have shifted to RASA based bot. we faced a lot of chalanges and have a wish list so it will be nice if we have contact to share our impressions on RASA and also to know your road map
post by Nishet on Apr 22, 2019
hey @hossa i check your bot and it was awesome, you do great work. Can you Share your github repo ya tutorial for that because it’s help to someone.i waiting your favorable reply.Thanks
post by aherrera on Apr 24, 2019
dadecoza](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hey @dadecoza! Your setup looks great! Do you use dockerhub as part of that? I’ve seen that it’s not free for companies. Do you know if dockerhub is needed or there’s acutually a free cost alternative?
post by dadecoza on Apr 24, 2019
hi @aherrera,
We do not use dockerhub, well not to host our images. We store our code and Dockerfile in GitLab and build the images as required from there. ie.
git clone some-project
cd some-project
docker build . -t some-project
Hope that helps.
post by aherrera on Apr 24, 2019
Yes it helps, thank you! I need to make a similar setup but I’ve never used Docker or Jenkins. Is there any link you could point me to that would help replicating this setup? I would need to know the number of virtual machines/servers needed, the number of images/containers in each environment and the deployment process from dev to prod.
post by azizullah2017 on Apr 30, 2019
hossa95](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
can you share please share your data or stories ?
1 month later
post by hemamalini on Jun 3, 2019
hemamaliniRasa Certified Developer
dadecoza](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hi, I am using rest channel to connect to rasa server from my website. In order to support concurrent user request, I need to change the sender id? How rasa can support concurrent request without docker
post by dadecoza on Jun 4, 2019
in the REST request you will have to send the “sender” attribute with the message … ie …
{
"message": "hello world",
"sender": "hemamalini"
}
The sender can be anything so you could use the id you use to track the user session in your web app, or generate a guid when the page loads …
1 month later
post by AGi on Jul 15, 2019
souvikg10](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hi Souvik can you please share your LinkedIn profile. Will connect there and talk about our requirement.
post by alex38 on Jul 15, 2019
I am currently working on my master thesis, which deals with the topic of machine ethics in general and especially with ways to model ethical decisions in a machine-understandable manner. The task was to implement a chatbot that is able to collect information from users about ethical dilemmas and process them with regard to established ethical principles, namely deontology and utilitarism. This is for sure pretty challenging, as the structure of such a dialogue is rather complex. Also, as could be expected, the lack of training data is a prevalent problem. Serious testing requires some interest and familiarization with the topic and there are only a few people who took the time to look more closely into the bot and whose dialogues could be included in the dataset accordingly.
At the moment, there is a prototype running at https://ethicbot.com/ which operates with a rather static dialogue structure. Currently, and for the remaining time of my thesis, I’m evaluating the bot’s performance and collect aspects to improve in a future version. Many dialogues by test users are derailed pretty quickly due to wrong classification, or because the bot is not able to deal with dialogues that deviate from his known patterns and the corresponding stories. Still, sometimes the concept works and meaningful dialogues come about, especially when users get used to the bot’s conversational style and get a feeling for the type of input the model works best with.
Overall, implementing the bot with Rasa has been a precious experience which has taught me a lot about conversational AI, more than I can devote to the project in the remaining time. However, I am pretty sure that this won’t be my last project with Rasa as I have learned so much during development and also had a lot of fun.
post by erohmensing on Jul 15, 2019
@alex38 your project sounds really cool! And having real conversations with people is always the best way to develop your bot’s conversational skills, it’s always surprising how people try to interact with a bot in ways you don’t expect glad you had fun working with Rasa!
post by souvikg10 on Jul 17, 2019
souvikg10Souvik GhoshRasa Team
AGi](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
@AGi, you will find me here : https://www.linkedin.com/in/souvik-ghosh-aaa30470/
post by yaput on Jul 18, 2019
Migrated Bot-kit to RASA Stack(NLU and Core), and now it’s on production, it’s a healthcare chatbot, still facing some issue but, it’s better to make it happen and make it better with the time.
The main problem that i’m facing are:
- handle user’s chit chat whenever and anywhere within the stories
- need to understand more about rasa built-in channel, right now use python script to serve the websocket connection to the chat widget
- use tracker properly to get all useful insight ( use dialogue tracker only now, without event broker)
- tried embedding pipeline, but it is too slow because i need a better instance also, and the stories are too many (for this particular bot in my opinion)
12 days later
post by ccelotto on Jul 30, 2019
Hey everyone!
We launched our chatbot on Saturday (using Rasa of course ) Check it out on Messenger: m.me/jadelasvegas
Jade is a Las Vegas personal assistant that recommends restaurants and bars on the strip.
Our company is Black Tie, a Las Vegas bot startup. We would love to get the community’s feedback! We would also love to thank the Rasa team for making this possible. You guys are killing it and we will continue to support and leverage your stack in delivering the best Las Vegas chatbot.
Chris
Co-Founder | Black Tie
14 days later
post by sarankarthik on Aug 14, 2019
harshitazilen](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hallo Harshitazilen! Thanks for sharing it. It gave me hope that I can use Rasa NLU and Rasa core for building a bot in production. Could you explainn me whether you have customized opensource dataset to the supported format of the Rasa or else you have created your own stories to train NLU and Core?
post by harshitazilen on Aug 14, 2019
I created my own intents and stories to train NLU and Core.
post by sarankarthik on Aug 15, 2019
Thanks for the Information!
post by sarankarthik on Aug 15, 2019
ccelotto](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hi ccelotto!
It is really interesting that your resataurant and Bar recommendation bot using Rasa is on production.
I am new to Rasa and I have to accomplish my Project using Rasa.
Can you explain me whether you have trained Rasa NLU and Core by creating your own intents and stories or else have you customized any open source datasets to achieve it? Can you tell me Average number of turns in the conversations you have used in the stories to train Rasa Core? Have you used templates for NLG or else have you used pre-trained NLG models to create natural Responses?
2 months later
post by gsagarvarma on Oct 7, 2019
dadecoza](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
How do we productionize Rasa? How to scale our Rasa application to handle 1000 of concurrent users? Can you please share the techniques and approaches for scaling the bot to handle more than 1000 concurrent user. Thanks in advance
1 month later
post by aayushgoel96 on Nov 14, 2019
As mentioned by @dadecoza, You just need to maintain and pass the session ID as sender key’s value in the request JSON. Session ID are generally maintained by your Web app.
post by usst on Nov 20, 2019
Juste](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hi Juste！I found a problem when I used the jmeter software to test the interface.Returned to empty in my custom action（To eliminate external factors）.When I increase the thread, the response time will increase.For example: Number of Threads = 2、run time = 30minutes
\ image1451×589 50.8 KB](https://europe1.discourse-cdn.com/flex013/uploads/rasa/original/2X/a/aa66ae0f9b495896e0138c5315baf99eb24afecf.png "image")
Number of Threads = 5、run time = 30minutes
\ image1480×588 54 KB](https://europe1.discourse-cdn.com/flex013/uploads/rasa/original/2X/9/90dab013e41c7eaa081ae5a647699df5b8912ddc.png "image")
I tried to increase the Number of threads to 100 running time of 8 hours, and the average response time is about 25 seconds.So,I want to know if rasa has requirements for Number of threads，and how many users can be accessed by rasa at the same time? Thanks！
post by pausali.biswas on Nov 26, 2019
Hi @Juste,
Same question again, if there is any documentation for the below- I have created chat bot on local and want to test it on production. Could you please suggest some documentation about to hosting bot on production. I am very new on development and it will be very helpful for me. Thank in advance
post by shubham3582 on Nov 26, 2019
@Juste This will be very helpful for all of us. Meanwhile I have few suggestions:
- Run Rasa on high performance server/machine (minium 4 cores and 16GB RAM)
- You can run duckling on separate server (but need to change conf file of rasa bot
- Also actions servers can be hosted on separate server(need to check if it works), but rasa must be installed on that machine.
- Frontend of your bot can be connect via socket or webservices (REST based).
post by shubham3582 on Nov 26, 2019
ccelotto](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hi, your bot works very fast. But as per my experience you need to add more training data to this bot. If possible can you share more details.
post by pausali.biswas on Nov 27, 2019
@shubham3582 can we have a multi Chats functionality open for the same, as in i have installed rasa and currently it is running locally fine, but only for a single session, so you know how can we enable multiple chats.
post by shubham3582 on Nov 27, 2019
Do you want to run multiple chatbot instances or you want to run multiple chats for different users ? Multiple chatbot instances may not possible to run with rasa community edition. For running the chatbot for different users can be achieved. Your application session/bearer token needs to taken into account while developing your bit.
post by pausali.biswas on Nov 27, 2019
@shubham3582 to run multiple chats for different users, this is what i want. "our application session/bearer token needs to taken into account while developing your bit."Can you detail me more about this, is there any configuration changes that needs to be done on the RASA side, is this already present in RASA.
post by shubham3582 on Nov 29, 2019
As long as you give each user a unique sender_id in the requests you make to your input channel, this shouldn’t be an issue – each person will have their own unique conversation with the bot.
post by pausali.biswas on Nov 29, 2019
Can you share with me how can we do this, any configuration changes required or we have to code the same?
post by amir on Dec 1, 2019
amirAmir KhanAdvanced Rasa Certified Developer
@shubham3582 I too want to know how to handle chat with multiple users at the same time, you are talking about making request with unique sender_id. you are also talking about application session/bearer token. I don’t know how to achieve this. If you can help me out it will be a great support from your side. Is it something to do with tracker.sender_id inside actions.py?
post by shubham3582 on Dec 4, 2019
I might help you better, if you elaborate the exact scenario.
post by amir on Dec 7, 2019
@shubham3582 suppose I have a rasa bot running on my local and connected with slack group channel. In group there are multiple users who are chatting with rasa bot at the same time. There is problem that rasa bot can’t handle multiple users at same time. Is it due to my rasa bot running local that’s why it can’t handle it. If I deploy me bot to cloud using aws or gcp or Azure services will bot handle multiple users at same time.
Is there a way to handle each user in that situation?
19 days later
post by TQuy on Dec 26, 2019
I want to run multiple instances of Rasa on different servers so that we can handle 100s conversations at the same time. How can I do that?
14 days later
post by amir on Jan 10, 2020
@TQuy I think… Its related to how you deploy rasa… I think u deploy through docker with kubernetes on server cloud platform will help you to scale the rasa… I haven’t tried that yet… what u think in this
15 days later
post by Manju_RASA on Jan 26, 2020
Hi, Is there anyone in Asia who has deployed a bot for any bank? I love to you hear your story about building a banking chatbot using RASA.
19 days later
post by bayesianwannabe on Feb 14, 2020
Do you guys know about some companies using it in Brazil?
19 days later
post by lgabs on Mar 5, 2020
lgabsLuan FernandesAdvanced Rasa Certified Developer
Hi @bayesianwannabe rs. I’m brazilian and I’m working on a chatbot (look for my name at introduction thread on this forum). The best case I’ve found was Lappis, built by a group from UnB ( GitHub - lappis-unb/rasa-ptbr-boilerplate: Um template para criar um FAQ chatbot usando Rasa, Rocket.chat, elastic search). There is a telegram group with interested people either, take a look at the repo.
post by bayesianwannabe on Mar 10, 2020
Hey Luan,
Thanks for your kind reply. Do you have a chatbot using Rasa in production? Can you tell more about the number of intents and your greatest challenge?
Lappis is the biggest case I know and the boilerplate project seems neat (I still need to test, but I am now more dedicated in a design/model prototyping part of the project), but I wonder if there are big companies on Brazil using this technology like banks or big industries.
post by skjainmiah on Mar 17, 2020
I’m waiting for correct answer regarding bots in production. I built a FAQ bot and my organisation is not happy with it as it cannot be deployed in website and I said I will do that but I’m not getting exact resource in the rasa docs so please help me regarding this. If I want to deploy in a website what is the exact process for it please share it @Juste. Thank you
1 month later
post by tacopanaya on Apr 19, 2020
Hello @skjainmiah. I just finished deploying RASA in a WordPress website so I’ll show you what I did and maybe this could help:
I’m running my RASA chatbot using ngrok, so I have my ngrok URL opening a specific port to the internet (in this case, port 5005).
I added the following code to credentials.yml, to be able to use socketio:
socketio: user_message_evt: user_uttered bot_message_evt: bot_uttered session_persistence: true
I used webchat and deployed in my frontend using the tag you can find here: GitHub - botfront/rasa-webchat: A feature-rich chat widget for Rasa and Botfront
I changed “socketUrl” with my ngrok URL.
And that’s it. Bot is working in frontend.
Hope this can help.
post by skjainmiah on Apr 19, 2020
harshitazilen](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hi @tacopanaya. It was good to hear you deployed on wordpress. How many people can access your chatbot at a time? does ngrok running continuously in the backend?
post by tacopanaya on Apr 20, 2020
Hi @skjainmiah. We are still in dev alpha stage. Up until now we were using Telegram for testing purposes but we faced some issues with button generation that weren’t present when we tested in webchat. As we are probably going to move to a mobile app in the future for full control over what we are creating, we decided moving to a webchat was a good next step. So we open ngrok before testing and whenever we need to show it to someone. We are probably going to go for an Amazon EC3 server in the future whenever we decide to open for public testing.
However, as far as I’ve seen in Rasa documentation here at the bottom, ngrok limit is set to 40 connections/minute.
post by viktomer on Apr 25, 2020
I want to know if a rasa core model can handle 2k-3k concurrent request on a 15gb ram server?
1 month later
post by KamyarGhajar on May 26, 2020
Hi, We are using Rasa Open Source for NLU. The http response for sequence tagging and intent extraction (/model/parse endpoint) is very slow (about than 2000ms on average and about 22 rps that is very low) with so many timeouts. The rasa 1.10.1-full docker is used on an OpenShift based cloud with 2Gi of ram and 2 CPU cores and horizontal scaling does not help so much. We have tseted it in a local machine with 64GB of ram and 12 i7 cores, the results are the same. What is the problem here that the rasa server is so slow for model parsing? Tests are done with wrk tool with 20 threads and 30 connections. With higher connection numbers than 20 or 30 it will even make all the requests to time out. Lower than 20 or 30 connections it is as I have described.
post by GeovanaRamos on May 27, 2020
bayesianwannabe](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hi @bayesianwannabe, take a look here.
post by bayesianwannabe on May 27, 2020
Nice! Thank you, Geovana. It’s always good to check new local cases using the Rasa framework.
1 month later
post by Controller-notai on Jul 7, 2020
raghibomar](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hi can you guide me on the cost of having this? I am trying to pitch it internally in my company and require a basic cost. I would have similar numbers as you defined earlier.
Also if you could let me know if you are using AWS or some other platform
1 month later
post by shang on Aug 11, 2020
Hi I’m new to Rasa here. Just wondering if it is possible to integrate CRMs such as Marketo with the chatbot and send information back and forth to personalise the chatbot experience for the user? It will be cool if the user gets a personalised message when he/she uses the chatbot
22 days later
post by reddyakilesh22 on Sep 3, 2020
Hello, I’m new to rasa and I have finally made a simple chatbot and now I want to deploy it on a website as of now I have made a simple blank website using this video as source Rasa chatbot using Webchat (Rasa bot part 3) - YouTube but unable to to get the chat widget. Can someone help me what might be the problem ? And In that video the src = “ https://storage.googleapis.com/mrbot-cdn/webchat-0.5.3.js” but now the webchat github shows src=“ https://cdn.jsdelivr.net/npm/rasa-webchat/lib/index.min.js” want to know whats the difference between those and I have tried both which didn’t work. @Juste Please help me out.
post by mloubser on Sep 3, 2020
Follow the instructions on the webchat github - they are up to date where their own URL is concerned
1 month later
post by robynbitner on Oct 10, 2020
robynbitnerRasa Certified Developer
TQuy](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
I would like to hear an answer for this as well!
post by poojadeshmukh on Oct 16, 2020
poojadeshmukhPoojaRasa Certified Developer
dadecoza](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hi @dadecoza,
Can you please help how did you productionize it?
- I have customized UI written using AJAX and JavaScript. It runs very well on my local with command.
- When we deploy Rasa on production, do we need run and keep the actions server command (rasa run actions) and the command prompt to run the UI continuously ON? What is the way to do it on Server?
- Please share any video/documents on productionizing Rasa.
Thanks
post by poojadeshmukh on Oct 16, 2020
TQuy](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
@TQuy, can you please help to know how to deploy it on windows server ? After deployment how to make it available to users? Generate URL? How to keep actions server and rasa server running? Please help.
23 days later
post by RasaRome on Nov 9, 2020
Good Morning,
kindly does anyone know expert RASA personnel in the Rome area ? thank you so much
post by Yash_kals on Nov 9, 2020
I’ve created one chatbot with all HR functionality for my company which is included daily punch in/out, timesheet, total hour count, add employee for HR, HR can show any employee daily data,HR can show who had done overtime with only one keyword ‘overtime’ and employee can used this chatbot as personal assistant like any user can set reminder at any time with reminder message and also cancel reminder features and many more functionality i’ve added in my chatbot which is created in Rasa.
9 days later
post by kiranbeethoju on Nov 19, 2020
kiranbeethojuRasa Certified Developer
It’s all about a healthcare bot…
I’m building a chat bot with more than 150 intents, 2 lakh entities, 1600+ stories, 16700+ NLU training examples which are prepared manually, the bot is now working well, it’s responding to any query which is near match of 16700+ training line items, I also used external API calls in actions. Customised bot UI and added carousel with button which can redirect user to respective URL. But the challenge here is I’m not able to get localStorage data of a user from a browser when user is logged in.
Your guys are not explaining anything about lookup tables in your workshops, lookup table feature is awesome why don’t you help us in fitting that lookup tables correctly in NLU file?
And thanks to all Devs and contributors. Here are a few snaps…
\ IMG_20201119_150802523×1078 218 KB](https://europe1.discourse-cdn.com/flex013/uploads/rasa/original/2X/d/dd1e4076d9e67b9f1af0bf07b7c0f6d8dcada81c.png "IMG_20201119_150802")
\ IMG_20201119_150630523×1028 149 KB](https://europe1.discourse-cdn.com/flex013/uploads/rasa/original/2X/9/94ff0fabc229f3b346b124603e7effe24221fbf5.png "IMG_20201119_150630")
post by Lalit_Chaturvedi on Nov 25, 2020
Hello ,
I was finding the way to deploy rasa on windows without using docker , but i didn’t get anything anywhere then i find a customized way to do such thing.
- Install RASA
- create custom actions
- Create windows service to start and stop RASA/RASA action server so that it should be easy to maintain server.
- Train rasa model using the csv by updating YML files using the custom action itself.
- create a html and using rest i connected it with rasa server.
How it work as https url ?
- make your rasa server localhost url https using apache virtualhost entry.
- In the same way you can host your ui URL too.
For me it is working fine.
please let me know any better way if it is available.
Thanks, Lalit Chaturvedi
1 month later
post by darich10 on Dec 29, 2020
Hello everyone.
I have already run in production a few chatbots, and now I am starting to do some freelance job.
But I don’t now how to do an hour estimation project. Anybody has any advice? How do you usually do a proposal document?
Any advice is wellcome. Thanks in advance.
1 month later
post by octobrtech on Jan 30, 2021
1
I’m building a commercial AI assistant using Rasa and I’m looking for AI assistant/chatbot specialists who have good experience with Rasa stack to help me. Please contact me
post by prakashr on Feb 2, 2021
prakashrRasa Certified Developer
hemamalini](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
@hemamalini Rasa can only wait for 5 seconds while it was in some actions running so you can have a seperate flask app to maintain it
15 days later
post by Suyog-23 on Feb 18, 2021
criada.in is using rasa to build their chatbots We will release our bot very soon, I’ll link it here
15 days later
post by goodyonsen on Mar 5, 2021
Hey,
I’m new in this forum, to the topic, and to the programming itself. It’s been only few months since I started learn to code (python only). I took some online certification courses and ended up having heard about the chatbots, the idea of which I liked very much. So, in my understanding I first need to install the open source -not the x- to be able to work with the chatbot I created using Rasa packs which I will be configuring based on my own preferences. I’m wondering if there’s anything else I need to know before starting. I appreciate the response.
post by vi.kumar on Mar 12, 2021
Juste](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hello Juste, We developed an AI chtabot for our firm. It is in beta testing phase. We have used both RASA NLU and RASA CORE. Rasa version - 1.10.12
Our front end is - Company Messenger App
We have hosted a web server on Django framework as a bridge between Company Messenger Server and Rasa server.
The amount of users i am handling at one time is still not efficient.
Can u tell me how shall i deploy it so that i can handle 200 users at a time.
Currently what i have done i am telling u- I started Actions Server , Rasa models server From webserver which is hosted on django i am calling rasa models server via REST api.
5 months later
post by kiranbeethoju on Jul 26, 2021
kiranbeethojuRasa Certified Developer
My bot is running in production from last 3 months I can say it is a massive bot trained on 90k training data, 150+ intents, around 2lakh entities.
You can find it on https://www.callhealth.com
myHealthCHat V1 CallHealth's chatbot in production - Using Rasa in backend - YouTube
){kind=link}
\ Screenshot_2021-07-27-06-17-24-607_com.android.chrome1080×2160 241 KB](https://europe1.discourse-cdn.com/flex013/uploads/rasa/original/3X/c/b/cb28b01f3991621fabbb52fbef93e9e3b896902a.jpeg "Screenshot_2021-07-27-06-17-24-607_com.android.chrome")
\ Screenshot_2021-07-27-06-18-07-845_com.android.chrome1080×2160 218 KB](https://europe1.discourse-cdn.com/flex013/uploads/rasa/original/3X/3/5/35ad2ad2c67e33e6fbc8cfbf9a15d55bfaff3bb3.jpeg "Screenshot_2021-07-27-06-18-07-845_com.android.chrome")
1 month later
post by Soltares on Aug 30, 2021
My family runs a small software business and we put Rasa into production about two months ago to help field FAQ related support questions. At the time of writing we are up to about 2,813 examples across 296 intents and it’s working very well! Shout-out to @rctatman, @Juste, @koaning, and the Rasa team for your amazing suite of helpful content which enabled us to get up and running.
Our bot’s name is Juliet and she can be found here: http://juliet.affirmatech.com
The model is primarily one very large ResponseSelector centered around the domain of our Amateur Radio logging software. There is one story, however, for collecting a user’s information via a Rasa form in order to resend their registration details in the event it is lost or forgotten.
I wrote a custom web interface using Python and Flask to serve both as a means for user interaction as well as a quick way for us to classify examples into their corresponding intent. Rasa X would have worked just as well, but I liked the challenge
\ image1920×1080 321 KB](https://europe1.discourse-cdn.com/flex013/uploads/rasa/original/3X/0/b/0b7b53c3f7befb9a3b1f4673c51f3fde0527796f.png "image")
For fun, I also wrote a quick Discord bot interface and its come in handy.
Thanks again to the Rasa team for making Juliet possible!
post by rctatman on Aug 30, 2021
rctatmanRachael Tatman Enthusiast
That’s great to hear! grats on the launch
post by koaning on Sep 6, 2021
koaningVincent D. Warmerdam Rasa Certified Developer
Soltares](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
That’s super cool! Thanks for letting us know!
3 months later
post by JackWhite on Dec 3, 2021
Hey! Now I’m working with implementing dragonfly technology to my bot. I have a telegram channel that covers space topics and this one will be implemented to the bot so the users can see places as photos after typing commands. I wonder is there any good advice or tips for implementing such a thing? Now I am halfway through but maybe someone did something like that before and I would not mind hearing about your experience. Thanks.
1 month later
post by baval on Jan 13, 2022
bavalBavalpreet SinghRasa Certified Developer
Link to our bot in production - IDP | Home Page
Background
The “India Data Portal” is a one-stop open-access portal for journalists to access, interact with, and visualize information, data, and knowledge related to agriculture and financial inclusion, while also aiding other beneficiaries – researchers, students, policymakers, administrators, NGOs, and entrepreneurs. The portal contains a data repository with processed and documented public datasets on related themes.
The Challenge
To resolve queries related to specific datasets, helping users to create different visualizations in a guided manner, in addition to not only solving generic queries about the portal and its usage but also to provide functionality in order to get a user’s feedback through a specifically designed form. Furthermore, this support needs to be multilingual.
The Solution
Our chatbot provides a Multilingual solution to the IDP team which can handle the user queries(generic and specific) and guide them on how to utilize the portal for their specific tasks. This helped the journalists to understand and make different visualizations by using different datasets from the portal, providing information regarding the fellowship program and various ongoing masterclasses and training sessions. Furthermore, users can also provide feedback by utilizing the functionality of a form.
1 month later
post by Hibran on Feb 24, 2022
Awesome! Is it currently running in production? I would love to try it out if there is a chance!
12 days later
post by ashok5451chicha on Mar 9, 2022
Juste](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
@Juste i am just trying move the developed bot to production can you let me know the practical server,hardware,os,memory requirements for 10000 users concurrency at once.
6 months later
post by mccurleyfries on Sep 5, 2022
has anybody integrated with a voice service? I am wondering what issues you had from a Customer Experience lens (eg: bot interrupting) and how they were overcome
3 months later
post by daniel12 on Nov 24, 2022
Hello, I have been involved in the development of conversational agents for a long time both at the university and now in my company. Namely, I started making conversational agents with version 2.0, at that time I wasn’t monitoring my memory that much. Now that I’m working for myself and we’re making a product for one company that includes speech technologies related to a robot and we have 50 conversational agents built for them, I’m interested in how, if there is any possibility, we can reduce the consumption of models. I am using the default pipeline and rasa version 3.3. I wonder if by changing the batch size I can change this. Currently, each model consumes between 800MB-1GB RAM depending on the size of the model. I also have a question, because it seems from my testing that the learning of the rasa models takes place on the cores, and I’m wondering if I can use some command to limit this or choose how many cores I want the rasa to occupy during training or startup. If, of course, anyone knows the answers to these questions, I would be very happy and maybe clarify some things for me.
Kind regards, Daniel
3 months later
post by mindseye on Feb 27, 2023
baval](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Hi Bavalpreet, Can you pls share what ML models did you use to support multiple languages?
Thanks!
1 year later
post by pranali05-ai on Apr 4, 2024
Hi Everyone,
I have used Rasa to build a bot for product support, where bot is specifically being used to collect user details and store it for the reference of relevant person to solve an issue. Rasa works fine at its minimal but when requirement grows it lacks in handling the complex things, for example differentiating between two similar inputs of different entity or intent. And many more other issues are there, I guess its challenge of open-source module.
5 months later
post by nour on Aug 17, 2024
neerajb1](https://forum.rasa.com/t/bots-in-production/520 "Load parent post")
Do you have a solution to deploy your bot plz if you have give me the main steps
Reply
Related topics
Topic list, column headers with buttons are sortable.
Loading complete
We collect and process your personal information for the following purposes: Visitor Statistics, Browsing Behavior. Learn more...
OKDecline