Question 1

What is a typical d²l consulting engagement?

Accepted Answer

There is no "typical" engagement. Team d²l customizes services based on where clients are in their data journey. The consulting team is handpicked for each engagement, as each member of Team d²l has unique expertise.

Services often include:
Design and implementation of data collection infrastructure (e.g., app data ingestion, text scraping, lead generation forms)
Automation of data processing, transformation, and warehousing
Optimizing data storage for process efficiency - anything from contact data in a brand new CRM to millions of rows stored on a cloud
Designing dashboards and reports with guidance on identifying key performance indicators
Providing executive-level insights from complex data analysis
Training, documentation, and hiring support

For specific examples, see Project Examples on the Consulting Services page.

Question 2

Why did d²l build Alkemy?

Accepted Answer

Alkemy started as a structured framework designed to reduce the common risks in ML projects - things like unreproducible results, fragile pipelines, and messy code. It was primarily a tool for Team d²l to use across consulting engagements to support consistency, reliability, and scale. As the team continued to use Alkemy, it became evident that client teams could benefit from direct access to the platform as a dependable foundation for building production-ready data science workflows. This realization led Team d²l to focus on hardening and enhancing Alkemy as a robust, user friendly product.

Question 3

What is Alkemy, where does it fit, how does it work, and how would my team adopt it?

Accepted Answer

Alkemy is a Python framework that takes the practical challenges and pitfalls of machine learning (ML) projects head on. It delivers accessibility and flexibility at every stage of the ML lifecycle. Alkemy was built for data science teams who need to deliver. It gives you consistent patterns, provides plumbing that dramatically reduces boilerplate code, and then it gets out of your way. You choose the right tools for the job, and you can build with any library you want. The best way to adopt Alkemy is to try it out with a new ML problem or pick an existing project and compare it side by side.

Question 4

What’s the average time it takes to start modeling with Alkemy?

Accepted Answer

You can familiarize yourself with Alkemy and get up and running with an initialized project within a couple of hours.

Question 5

What are some example applications of Alkemy models?

Accepted Answer

Alkemy can be been applied to a wide variety of predictive modeling scenarios including trading, sports betting, text classification, sentiment analysis and much more.Team d²l’s vision is for Alkemy to serve as a flexible, general-purpose framework that can support predictive use cases across industries.

Question 6

What skills do I need to be successful using Alkemy?

Accepted Answer

Success with Alkemy comes from a combination of technical skills and good judgment. Users should be comfortable with Python development, SQL/data handling, and machine learning fundamentals, along with working in configuration files and managing reproducible workflows using Git and DVC.  When working with an AI coding agent, users can rely on the agent for implementation, but should still be able to clearly define problems, understand their data, review generated code and configs, and evaluate results - combining domain knowledge with thoughtful oversight of the end-to-end Machine Learning (ML) process.

Question 7

What are the system requirements for running Alkemy?

Accepted Answer

Operating System: macOS
Memory: 16 GB RAM or higher
Processor: Apple Silicon (M1 or newer), MacBook Pro–class performance (Late 2020 or later)

Question 8

What does the onboarding process look like to start using Alkemy?

Accepted Answer

Our onboarding is a guided process where we help you set up your first project, define evaluation metrics, and run initial model comparisons. This typically includes:

Structuring your first project (datasets, metrics, models)
Setting up evaluation workflows
Training on how to compare models and make decisions
Configuring agent interactions and guardrails

This can take a few sessions over 1–2 weeks, depending on complexity.

Question 9

Do I have to use d²l consulting to use Alkemy?

Accepted Answer

In short, no.  Team d²l will provide access to the latest release, as well as the developer site, which includes a wealth of information, including key concepts, relevant guides and references, installation instructions, and a tutorial to help you gain practical insights and understanding.

Fill out the contact form to gain access and get started.

Question 10

How is Alkemy pricing structured?

Accepted Answer

Our pricing is simple and predictable. We use a subscription model based on number of users along with optional training packages:

Pro and Team plans are priced per user per month
Onboarding and training are one-time packages to help you get started quickly

FAQ