Over the past few months, I’ve been diving deep into Mojo 🔥, exploring its potential as a high-performance language that blends Python ergonomics with systems-level power. Along the way, I’ve found a number of excellent YouTube videos that have helped me better grasp both the language and the broader Modular vision for high-performance AI computing.
If you’re also exploring Mojo, these talks and demos are well worth checking out. They offer a mix of technical depth, vision, and practical examples—from GPU programming and PyTorch integration to community updates and deep dives with Chris Lattner, Mojo’s creator.
Here are some highlights that stood out to me:
-
From Swift to Mojo and High-Performance AI Engineering – A wide-ranging conversation with Chris Lattner on how Mojo builds on lessons learned from LLVM and Swift.
-
November 2025 Modular Community Meeting: 25.7 Release & Mojo 1.0 Roadmap – These meetings provide great snapshots of the Mojo ecosystem evolving, showcasing community projects and tooling updates.
-
High Performance AI in the Real World with Modular – An overview of Modular’s roadmap and the real-world workloads shaping Mojo’s design.
-
Mojo 🔥 - Is it Python's faster cousin or just hype? PyCon DE & PyData Berlin 2024 – An engaging introduction to Mojo for Python developers, with an objective look at the hype vs. potential.
-
Mojo GPU Puzzles Tutorials (Learn GPU Programming with Mojo 🔥 GPU Puzzles Tutorial - Introduction, Puzzle 01: Map, Puzzle 02: Zip, Puzzle 03: Guard) – A fun, hands-on series for learning GPU programming in Mojo through progressive puzzles and challenges. Note: Only the first three puzzles in the series have been released as of December 12, 2025. For the full interactive tutorial experience, check out the GPU Puzzles site.
-
Mojo - First Impression [Programming Languages Episode 29] – Mike Shah's view on Mojo over time, contrasting early impressions in this video and later with a 2025 revisit.
-
Mojo + PyTorch: A Simpler, Faster Path to Custom Kernels – Practical insight into using Mojo to simplify custom kernel development on top of PyTorch.
Together, these resources provide a balanced mix: visionary discussions, technical demos, and relatable walkthroughs. These videos make the Mojo learning curve smoother. If you’re following my Mojo series on DataBooth, you’ll likely see ideas from some of these pop up over time, as they’ve been instrumental in shaping how I think about the language’s design and direction.
Appendix: Mojo Video Watchlist
This list collects the talks and videos that have been especially helpful for my own learning so far; it is not intended to be exhaustive, and the Modular YouTube channel and broader Mojo ecosystem are evolving rapidly with new content and events.
-
Long-form interviews and vision:
-
Modular community and roadmap:
-
GPU programming and MAX:
- Learn GPU Programming with Mojo 🔥 GPU Puzzles Tutorial – Introduction (Modular)
- Mojo 🔥 GPU Puzzles Tutorials – Map, Zip, Guard (Modular)
- Mojo GPU Compilation 🔥 (Modular)
- Next-Gen GPU Programming: Hands-On with Mojo & MAX (Modular)
- Chris Lattner on High Performance AMD GPU Programming with Mojo (AMD Developer Central)
-
Language comparisons and first impressions:
- Mojo 🔥 – Is it Python's faster cousin or just hype? (PyCon DE & PyData Berlin 2024)
- Mojo vs Python: Why the Next Decade of AI Won’t Look the Same (From Scratch Podcast)
- Choosing Your Language: Python or Mojo? (ArjanCodes)
- Mojo – First Impression and 2025 Revisit Parts 1 & 2 (Mike Shah): First Impression, 2025 Revisit Part 1, 2025 Revisit Part 2
As new releases land and the documentation, GPU puzzles, and community projects expand, the Modular and community channels keep adding fresh material: so treat this Appendix as a snapshot, as of the date of this post, of what has helped so far rather than a definitive directory.