The Open Virtual Assistant

Almond lets you interact with your services and accounts in natural language, with flexibility and privacy.

Get Almond


Open & Crowdsourced

Almond draws its power from the crowdsourced Thingpedia, an open collection of Web and Internet of Things APIs. Anyone can contribute support for the favorite service, with few lines of code and a handful of natural language sentences.

Thingpedia supports more than 40 different services, accounts and IoT devices.

Almond is free software, and is available on Github. You can download, distribute or modify Almond as you see fit.

The logos of the services supported by Almond

Unprecedented Expressivity

When the Bitcoin price reaches $20,000, search for a “Bitcoin” picture, and tweet it with caption “I am rich!”

monitor @bitcoin.get_price() on price ≥ $20000 @bing.image_search("bitcoin") ⇒, "I am rich!")

Almond is the world first virtual assistant to support compound commands in natural language.

In Almond, you can combine multiple services into a single command, with powerful filtering and passing of data from one to another. Do away with the complex UIs of IFTTT or Zapier - in Almond the power is ready for you to type.

Respects Your Privacy (and That of Your Friends)

Almond is the world first Communicating Virtual Assistant: through Almond you can interact not only with your services, but also access that of your friends, family and colleagues.

To protect your privacy, Almond allows you to set flexible access control policies, based on time of day, on your location, on the content of the request, or really on any information available to Almond. For example, your parents can monitor your security camera, only if you're not home, or only at a certain time of the day.

Finally, to guarantee your privacy, Almond runs all its actions and computation locally. If you so choose, you can have your own installation of Almond in your phone or your laptop. If you use Web Almond, your data is encrypted and stored in a secure enclave not accessible by anybody else.

A dad can access the security camera of his daughter, until certain conditions specified by her.The architecture of communicating virtual assistants.

Publications & Talks

Our Team

Prof. Monica Lam

Monica Lam is a Professor in the Computer Science Department at Stanford University since 1988. She received a B.Sc. from University of British Columbia in 1980 and a Ph.D. in Computer S cience from Carnegie Mellon University in 1987. She is the Faculty Director of the Stanford MobiSocial Computing Laboratory and a co-PI in the POMI (Programmable Open Mobile Internet) 2020 project, which is an NSF Expedition started in 2008.

Prof. Richard Socher

Richard Socher is the chief scientist at Salesforce and an Adjunct Professor in the Computer Science Department at Stanford University. Previously, he was the founder and CEO/CTO of MetaMind. He received his PhD in 2014 from Stanford University, and he currently teaches CS244N Natural Language Processing with Deep Learning.

Giovanni Campagna

Computer Science PhD student

Platform backend, natural language, Thingpedia and ThingTalk design

Michael Fischer

Computer Science PhD student

UX design, HCI and graphics

Mehrad Moradshahi

Electrical Engineering PhD student

Natural language understanding

Rakesh Ramesh

Electrical Engineering PhD student

Natural language understanding

Silei Xu

Computer Science PhD student

Systems, Thingpedia design, distributed system

Richard Yang

Computer Science MS student

HCI and graphics

Previous members of our team include Albert Chen, Zhiyang He, Jiaqi Xue, Aashna Garg, Jiwon Seo, Sadjad Fouladi and Reynis Vazquez. We thank them for their valuable support.