The Alchemist Blog

Helping You Build Intelligent Data Applications

APIs and Libraries: Web Application Building Blocks

Post Author Avatar

By: Zach Walchuk on March 27, 2015 at 11:03 AM

Considering I grew up to become an engineer, it probably comes as no suprise that when I was younger the biggest highlight of every birthday was a new LEGO set. Each set gave me new pieces for my collection, and with every new piece came the possibility to build something entirely different. By the time I moved on to more durable construction methods, I knew how to work with gears, pulleys, belt drives, drive trains, and of course, spring-loaded exploding walls.

Like building with LEGOs, putting together a web application is a matter of combining the right pieces in the right configuration. Once you have established the overall purpose of your application, try to break it down into functional units. If you can find an API or library for each function, then all you have to do is figure out what you need to connect the pieces. This allows you, the developer, to focus solely on building something no one has built before.

Read More

Topics: User Stories, Application Development

A Deep Learning Primer For Product Managers

Post Author Avatar

By: Kelly Taylor on March 24, 2015 at 12:55 PM

Most of my discussions lately with CTOs and Product Managers have ended with me emailing some links and info about how they can use deep learning in their SaaS applications. I’m a relative newbie to the space myself so I created a primer.

Read More

Topics: Deep Learning & Machine Learning, How-To

5 Tools for Analyzing Large Data Sets

Post Author Avatar

By: Devin Harper on March 19, 2015 at 2:00 PM

Harnessing and exploring large data sets has become increasingly important for businesses in every industry. There are many companies and services trying to make this a tenable problem, and still even more people munging together home-grown solutions to meet their specific needs. I thought it might be an interesting time to jot down my thoughts on five tools in the market today that make working with large data sets a little more doable.

Read More

Topics: Big Data Analysis, Application Development

What Can a SaaS Company Do With AI Right Now?

Post Author Avatar

By: Kelly Taylor on March 17, 2015 at 1:30 PM

Most of the SaaS apps we use everyday share similar attributes like inputting text, uploading files and integrating with other tools. This post is written based on my knowledge of SaaS apps like Basecamp,, PivotalTracker, Jira, ProdPad, Slack, etc.

Here are a few features that most SaaS apps could improve or introduce using the power of AI.

Recommend Things For Me To Follow

Following team members or things is a common feature in SaaS apps but in my experience there isn't much intelligence behind what I should follow.

Ex: "Based on User Stories typically assigned to you, here are a few others you may want to follow"


The SaaS app knows a lot about me from my identity to messages I exchange with my team to what I'm working on. I want to see sophisticated benchmarks that combine content analytics with other metrics to provide insights that I can use.

Ex: "These keywords occur the most in teams that work on more than 25% bugs every week."

Ex: "These topics are found in stories that have relatively larger stories point estimates than others."

Show Similar Images

When teams are collaborating, images get scattered and buried, especially in the software development / agile tools I've used. When I upload a wireframe or screen grab, the app should show me other similar images that have already been uploaded.

Ex: "Someone on your team has already uploaded this image"

Search Vectoring

As data grows and grows inside the SaaS apps we all use, search becomes more important as a navigation tool. Searching on my name to see a list of things assigned to me, searching on keywords to see grouped things and searching by thing name to navigate to that thing because it's easier than navigating to it in the UI are all common SaaS app patterns today. That's all good, but I want more. I want to search for abstract concepts and get relevant results back. I want to search for things and see results for images that contain those things. I want search results to display in intelligent order, not just alphabetically or by date.

Ex: Search for "First Ladies of the United States" and get results containing "Michelle Obama" even though she is not specifically mentioned in the thing returned.

Ex: Search for "Positive feedback on that thing" and see a list of things that have a positive sentiment.

Read More

Topics: Deep Learning & Machine Learning, Big Data Analysis

Tabelog’s Recommendation Site Uses Image Analysis to Make You Hungry

Post Author Avatar

By: Sonya Hansen on March 12, 2015 at 2:30 PM

What’s even more social these days than sharing a meal with friends, family or your partner? Maybe it’s how you go about deciding on what you want the experience to be. Tabelog is putting a new twist on making that choice with an online restaurant directory that helps connect hungry people with a meal they can remember. At Tabelog, you’ll find reviews by everyday diners, critical insights and images from professional food bloggers, and photos pulled from well-known social sites that combine to create a great virtual representation of a restaurant’s atmosphere and food.

Read More

Topics: Computer Vision & Image Recognition

The Flask Framework: Easy Web Application Prototyping in Python

Post Author Avatar

By: Zach Walchuk on March 11, 2015 at 11:42 AM

When I started learning web development, I quickly realized there were many pieces that needed to work together to make a complete application. I started with the client-side pieces - HTML, CSS and JavaScript. Learning these made it possible for me to create functional, dynamic web pages. Shortly after, I started working with Python, a server-side language. With Python I could easily access APIs, connect to a database and analyze my data, combining tools such as AlchemyAPI, MongoDB, and R. It soon became apparent that I would need a good way to connect my processes running on the server with the display functionality of the client. After a little research, I came across Flask.

Flask is one of many web frameworks available for Python. A web framework is a set of services and tools specifically designed to implement web applications by handling all of the nitty-gritty details behind the scenes. I started with Flask because it is a lightweight framework that requires minimal setup to get a prototype application up and running. While Flask may not have the full feature set of other frameworks like Django, the basic concepts are easy to understand. This makes it the perfect framework for developers new to web applications.

Read More

Topics: How-To, Application Development

Lessons Learned: How We Drive Innovation With Internal Hackathons

Post Author Avatar

By: Mackenzie Sedlak on March 10, 2015 at 1:30 PM

Here at AlchemyAPI, an IBM company, we strive to foster an environment where innovation and creativity run freely. However, it is difficult to fuel ingenuity and drive future thinking while continuing to meet our commitments to the business. As such, we allocated two workdays to host an internal hackathon. During these two days, we put aside our current projects and took the opportunity to ‘scratch an itch’ and explore passion projects.

Read More

Topics: AlchemyAPI, Hackathons

The Ultimate Glossary For Developers of Smart Apps

Post Author Avatar

By: Sonya Hansen on March 5, 2015 at 11:30 AM

Common Deep Learning and Data Analysis Terms Explained

Doctors talk about the “chief complaint.” You might hear “frac job” in a discussion between two oil and gas engineers. How about “shank?” It could mean a homemade knife, one of a golfer’s worst nightmares or the end of an evening.

Read More

Topics: Deep Learning & Machine Learning, Big Data Analysis

Enabling a Smarter Future: AlchemyAPI Acquired by IBM

Post Author Avatar

By: Elliot Turner on March 4, 2015 at 6:39 AM

Ten years ago, I founded AlchemyAPI with the goal of democratizing artificial intelligence to enable a smarter future. Today, I'm proud to announce the next step in this amazing journey. AlchemyAPI is joining IBM as part of its Watson unit, the recognized global leader in cognitive computing. 

Before we jump into the details of this announcement, I'd like to convey a sincere thanks to everyone who has supported AlchemyAPI over the years. To our customers, users, investors and advocates: Your passion helped make AlchemyAPI what it is today -- and we're just getting started.

Read More

Topics: AlchemyAPI

How Can A Product Manager Use Artificial Intelligence?

Post Author Avatar

By: Kelly Taylor on March 2, 2015 at 10:09 AM

As a Product Manager, I'm always looking for new technologies that enable a competitive advantage. In the recent past, we've seen trends like “social” and “mobile” be predicted and then dominate the landscape, the next version of this is "smart" or "using artificial intelligence" (AI). 

But you probably know this already. Awesome people like Shivon Zilis from Bloomberg Beta are writing about machine intelligence trends.

Sidenote: If you're new to the machine learning craze, read my deep learning primer for product managers.

Basically, AI for web and mobile apps means combining natural language processing (NLP) with computer vision capabilities to enrich large amounts of unstructured data. Then, doing something amazing with that data that only machine intelligence could possibly enable.

Read More

Topics: Deep Learning & Machine Learning

Subscribe to Email Updates

Download an API Key

Share AlchemyAPI's Blog