Bob Belderbos
@bbelderbos.bsky.social
Co-founder @pybites.bsky.social. Python developer & coach helping devs level up with project-based learning & modern Python. Start here → https://pybit.es
I love how we go deep into Rust core features in our Scripter to Rust cohort.
Code example that came up:
Code example that came up:
February 6, 2026 at 4:00 PM
I love how we go deep into Rust core features in our Scripter to Rust cohort.
Code example that came up:
Code example that came up:
TIL that #GitHub renders Mermaid diagrams natively in markdown files. Just use a fenced code block with `mermaid` as the language. 🚀
Works in READMEs, issues, PRs, and wikis.
No external tools needed, the diagram lives in your md, version-controlled with your code. 😍
Works in READMEs, issues, PRs, and wikis.
No external tools needed, the diagram lives in your md, version-controlled with your code. 😍
February 5, 2026 at 12:52 PM
TIL that #GitHub renders Mermaid diagrams natively in markdown files. Just use a fenced code block with `mermaid` as the language. 🚀
Works in READMEs, issues, PRs, and wikis.
No external tools needed, the diagram lives in your md, version-controlled with your code. 😍
Works in READMEs, issues, PRs, and wikis.
No external tools needed, the diagram lives in your md, version-controlled with your code. 😍
Spotted in a code review:
```python
if units == "f":
temp_str = f"{celsius}°F"
else:
temp_str = f"{celsius}°C"
# Later in another file...
units = console.input("Units? (c/f): ").strip() or "c"
# And in yet another place...
```python
if units == "f":
temp_str = f"{celsius}°F"
else:
temp_str = f"{celsius}°C"
# Later in another file...
units = console.input("Units? (c/f): ").strip() or "c"
# And in yet another place...
February 5, 2026 at 11:00 AM
Spotted in a code review:
```python
if units == "f":
temp_str = f"{celsius}°F"
else:
temp_str = f"{celsius}°C"
# Later in another file...
units = console.input("Units? (c/f): ").strip() or "c"
# And in yet another place...
```python
if units == "f":
temp_str = f"{celsius}°F"
else:
temp_str = f"{celsius}°C"
# Later in another file...
units = console.input("Units? (c/f): ").strip() or "c"
# And in yet another place...
Formatting timestamps with manual string slicing? That can be fragile.
Instead of formatting timestamps all over your code, keep the formatting close to the data (encapsulation).
Instead of formatting timestamps all over your code, keep the formatting close to the data (encapsulation).
February 4, 2026 at 7:00 PM
Formatting timestamps with manual string slicing? That can be fragile.
Instead of formatting timestamps all over your code, keep the formatting close to the data (encapsulation).
Instead of formatting timestamps all over your code, keep the formatting close to the data (encapsulation).
“I need a CLI tool. I’ll build a CLI tool.”
Then someone asked:
“Can I access this via a REST API?”
My answer used to be:
“No… I’d have to rewrite everything.” 😬
Then someone asked:
“Can I access this via a REST API?”
My answer used to be:
“No… I’d have to rewrite everything.” 😬
February 4, 2026 at 1:14 PM
“I need a CLI tool. I’ll build a CLI tool.”
Then someone asked:
“Can I access this via a REST API?”
My answer used to be:
“No… I’d have to rewrite everything.” 😬
Then someone asked:
“Can I access this via a REST API?”
My answer used to be:
“No… I’d have to rewrite everything.” 😬
Stop scattering `if/elif` checks across your code to manage workflow states.
If your feature has a real lifecycle (draft → review → approved → published), model it as a **state machine**.
If your feature has a real lifecycle (draft → review → approved → published), model it as a **state machine**.
February 2, 2026 at 4:16 PM
Stop scattering `if/elif` checks across your code to manage workflow states.
If your feature has a real lifecycle (draft → review → approved → published), model it as a **state machine**.
If your feature has a real lifecycle (draft → review → approved → published), model it as a **state machine**.
Your favorite #Python 🐍 libraries are lying to you.
• `pip install orjson` - that's Rust.
• `pip install polars` - Rust.
• `pip install cryptography` - Rust.
• `pip install pydantic` - the core is Rust.
• `pip install ruff` - Rust again.
• `pip install orjson` - that's Rust.
• `pip install polars` - Rust.
• `pip install cryptography` - Rust.
• `pip install pydantic` - the core is Rust.
• `pip install ruff` - Rust again.
January 30, 2026 at 3:24 PM
Your favorite #Python 🐍 libraries are lying to you.
• `pip install orjson` - that's Rust.
• `pip install polars` - Rust.
• `pip install cryptography` - Rust.
• `pip install pydantic` - the core is Rust.
• `pip install ruff` - Rust again.
• `pip install orjson` - that's Rust.
• `pip install polars` - Rust.
• `pip install cryptography` - Rust.
• `pip install pydantic` - the core is Rust.
• `pip install ruff` - Rust again.
Create GitHub issues faster from the CLI
Combine `gh`, `fzf`, and your `$EDITOR` to create issues without leaving the terminal 👇
This gives you fuzzy search for assignees and your familiar editor for composing the issue body—much faster than the web UI.
Combine `gh`, `fzf`, and your `$EDITOR` to create issues without leaving the terminal 👇
This gives you fuzzy search for assignees and your familiar editor for composing the issue body—much faster than the web UI.
January 30, 2026 at 1:00 PM
Create GitHub issues faster from the CLI
Combine `gh`, `fzf`, and your `$EDITOR` to create issues without leaving the terminal 👇
This gives you fuzzy search for assignees and your familiar editor for composing the issue body—much faster than the web UI.
Combine `gh`, `fzf`, and your `$EDITOR` to create issues without leaving the terminal 👇
This gives you fuzzy search for assignees and your familiar editor for composing the issue body—much faster than the web UI.
Stop writing scripts, start building systems.
January 29, 2026 at 8:00 PM
Stop writing scripts, start building systems.
"The borrow checker is fighting me!" 🥊
No, it's not. It's saving you from a 3 AM debugging session three months from now.
In Python, variables are references. You pass an object around, everyone shares it.
In Rust, by default, Ownership is exclusive.
No, it's not. It's saving you from a 3 AM debugging session three months from now.
In Python, variables are references. You pass an object around, everyone shares it.
In Rust, by default, Ownership is exclusive.
January 29, 2026 at 1:00 PM
"The borrow checker is fighting me!" 🥊
No, it's not. It's saving you from a 3 AM debugging session three months from now.
In Python, variables are references. You pass an object around, everyone shares it.
In Rust, by default, Ownership is exclusive.
No, it's not. It's saving you from a 3 AM debugging session three months from now.
In Python, variables are references. You pass an object around, everyone shares it.
In Rust, by default, Ownership is exclusive.
Watching 10 hours of video tutorials makes you a viewer, not a coder.
Most developers stay in "Passive Consumption" mode because it feels safe. You watch an instructor set up a pytest environment or configure a Docker container and think, "I get it."
Most developers stay in "Passive Consumption" mode because it feels safe. You watch an instructor set up a pytest environment or configure a Docker container and think, "I get it."
January 28, 2026 at 3:48 PM
Watching 10 hours of video tutorials makes you a viewer, not a coder.
Most developers stay in "Passive Consumption" mode because it feels safe. You watch an instructor set up a pytest environment or configure a Docker container and think, "I get it."
Most developers stay in "Passive Consumption" mode because it feels safe. You watch an instructor set up a pytest environment or configure a Docker container and think, "I get it."
Errors should not pass silently, so stop hiding them.
In Week 2 of our Pybites dev Rust cohort, we're tackling one of Rust's most powerful features: explicit error handling.
Here's the mental shift that changes everything:
In Week 2 of our Pybites dev Rust cohort, we're tackling one of Rust's most powerful features: explicit error handling.
Here's the mental shift that changes everything:
January 28, 2026 at 1:00 PM
Errors should not pass silently, so stop hiding them.
In Week 2 of our Pybites dev Rust cohort, we're tackling one of Rust's most powerful features: explicit error handling.
Here's the mental shift that changes everything:
In Week 2 of our Pybites dev Rust cohort, we're tackling one of Rust's most powerful features: explicit error handling.
Here's the mental shift that changes everything:
Can you spot the bug? 🐛
You’ve finally moved your logic into a proper src/ layout so you can structure your pytest tests properly.
You’ve finally moved your logic into a proper src/ layout so you can structure your pytest tests properly.
January 27, 2026 at 2:09 PM
Can you spot the bug? 🐛
You’ve finally moved your logic into a proper src/ layout so you can structure your pytest tests properly.
You’ve finally moved your logic into a proper src/ layout so you can structure your pytest tests properly.
Don't argue about formatting in code reviews. Let the machine do it.
If you are still manually fixing spaces or imports, you are wasting brain cycles.
Your `.pre-commit-config.yaml` should be your best friend.
If you are still manually fixing spaces or imports, you are wasting brain cycles.
Your `.pre-commit-config.yaml` should be your best friend.
January 27, 2026 at 10:08 AM
Don't argue about formatting in code reviews. Let the machine do it.
If you are still manually fixing spaces or imports, you are wasting brain cycles.
Your `.pre-commit-config.yaml` should be your best friend.
If you are still manually fixing spaces or imports, you are wasting brain cycles.
Your `.pre-commit-config.yaml` should be your best friend.
I just discovered a neat package to convert Markdown to HTML: markdown-it-py
I just used it to send a weekly markdown report, which now looks so much better :)
`markdown-it-py` made it trivial 👇
Works great with SendGrid, SES, or any email service that accepts HTML content.
I just used it to send a weekly markdown report, which now looks so much better :)
`markdown-it-py` made it trivial 👇
Works great with SendGrid, SES, or any email service that accepts HTML content.
January 26, 2026 at 5:42 PM
I just discovered a neat package to convert Markdown to HTML: markdown-it-py
I just used it to send a weekly markdown report, which now looks so much better :)
`markdown-it-py` made it trivial 👇
Works great with SendGrid, SES, or any email service that accepts HTML content.
I just used it to send a weekly markdown report, which now looks so much better :)
`markdown-it-py` made it trivial 👇
Works great with SendGrid, SES, or any email service that accepts HTML content.
Confidence comes from Shipping
There is no better feeling than `git push` -> `Deployed`. 🚀
Last week, one of our PDC Foundations students went from "I've never packaged an app" to having their CLI tool installable via `pip` - we love seeing these lightbulb moments. 💡
There is no better feeling than `git push` -> `Deployed`. 🚀
Last week, one of our PDC Foundations students went from "I've never packaged an app" to having their CLI tool installable via `pip` - we love seeing these lightbulb moments. 💡
January 26, 2026 at 8:02 AM
Confidence comes from Shipping
There is no better feeling than `git push` -> `Deployed`. 🚀
Last week, one of our PDC Foundations students went from "I've never packaged an app" to having their CLI tool installable via `pip` - we love seeing these lightbulb moments. 💡
There is no better feeling than `git push` -> `Deployed`. 🚀
Last week, one of our PDC Foundations students went from "I've never packaged an app" to having their CLI tool installable via `pip` - we love seeing these lightbulb moments. 💡
Agents are Interfaces, not just Chatbots
The real power of LLMs unlocks when you give them typed tools.
In Python, we can define a function and expose it to the model using Pydantic schemas.
The real power of LLMs unlocks when you give them typed tools.
In Python, we can define a function and expose it to the model using Pydantic schemas.
January 25, 2026 at 8:00 PM
Agents are Interfaces, not just Chatbots
The real power of LLMs unlocks when you give them typed tools.
In Python, we can define a function and expose it to the model using Pydantic schemas.
The real power of LLMs unlocks when you give them typed tools.
In Python, we can define a function and expose it to the model using Pydantic schemas.
Let's talk about Rust vs Python and type safety today.
Coming from Python, you might think of Enums as just a list of named constants: `Color.RED`, `Color.BLUE`, etc.
Coming from Python, you might think of Enums as just a list of named constants: `Color.RED`, `Color.BLUE`, etc.
January 24, 2026 at 11:00 AM
Let's talk about Rust vs Python and type safety today.
Coming from Python, you might think of Enums as just a list of named constants: `Color.RED`, `Color.BLUE`, etc.
Coming from Python, you might think of Enums as just a list of named constants: `Color.RED`, `Color.BLUE`, etc.
The difference between a Junior and Senior dev isn't syntax. It's this.
Most developers can write a script to fetch data from an API. They get it to work, then they move on.
Most developers can write a script to fetch data from an API. They get it to work, then they move on.
January 23, 2026 at 3:50 PM
The difference between a Junior and Senior dev isn't syntax. It's this.
Most developers can write a script to fetch data from an API. They get it to work, then they move on.
Most developers can write a script to fetch data from an API. They get it to work, then they move on.
You don’t learn by watching. You learn by breaking things. 💡
We've spoken with developers who’d watched multiple 10+ hour Udemy courses.
They knew the theory: Django, Docker, AsyncIO.
We've spoken with developers who’d watched multiple 10+ hour Udemy courses.
They knew the theory: Django, Docker, AsyncIO.
January 23, 2026 at 12:00 PM
You don’t learn by watching. You learn by breaking things. 💡
We've spoken with developers who’d watched multiple 10+ hour Udemy courses.
They knew the theory: Django, Docker, AsyncIO.
We've spoken with developers who’d watched multiple 10+ hour Udemy courses.
They knew the theory: Django, Docker, AsyncIO.
Why you probably don't need React 💡
"I need a React frontend for my SaaS."
Do you? Or do you just want a smooth UI?
The complexity cost of a decoupled frontend is massive:
- State synchronization
- JWT/Auth handling
- API serialization overhead
"I need a React frontend for my SaaS."
Do you? Or do you just want a smooth UI?
The complexity cost of a decoupled frontend is massive:
- State synchronization
- JWT/Auth handling
- API serialization overhead
January 22, 2026 at 9:00 PM
Why you probably don't need React 💡
"I need a React frontend for my SaaS."
Do you? Or do you just want a smooth UI?
The complexity cost of a decoupled frontend is massive:
- State synchronization
- JWT/Auth handling
- API serialization overhead
"I need a React frontend for my SaaS."
Do you? Or do you just want a smooth UI?
The complexity cost of a decoupled frontend is massive:
- State synchronization
- JWT/Auth handling
- API serialization overhead
We just shipped a new SQLModel Learning Path on the Pybites Platform:
🔹 Learn the Python ORM to focus on next
🔹 Get type-safe models with Pydantic-style, high-performance validation
🔹 See how cleanly SQLModel plugs into FastAPI for real-world APIs
🔹 Learn the Python ORM to focus on next
🔹 Get type-safe models with Pydantic-style, high-performance validation
🔹 See how cleanly SQLModel plugs into FastAPI for real-world APIs
January 22, 2026 at 6:32 PM
We just shipped a new SQLModel Learning Path on the Pybites Platform:
🔹 Learn the Python ORM to focus on next
🔹 Get type-safe models with Pydantic-style, high-performance validation
🔹 See how cleanly SQLModel plugs into FastAPI for real-world APIs
🔹 Learn the Python ORM to focus on next
🔹 Get type-safe models with Pydantic-style, high-performance validation
🔹 See how cleanly SQLModel plugs into FastAPI for real-world APIs
🐍 Pythonista → 🦀 Rustacean?
I’ve built a Rust Cohort for Pythonistas with Jim Hodapp, a 6-week, project-driven journey where you’ll:
Build a real JSON parser, in Rust 🏗️
I’ve built a Rust Cohort for Pythonistas with Jim Hodapp, a 6-week, project-driven journey where you’ll:
Build a real JSON parser, in Rust 🏗️
January 22, 2026 at 7:14 AM
🐍 Pythonista → 🦀 Rustacean?
I’ve built a Rust Cohort for Pythonistas with Jim Hodapp, a 6-week, project-driven journey where you’ll:
Build a real JSON parser, in Rust 🏗️
I’ve built a Rust Cohort for Pythonistas with Jim Hodapp, a 6-week, project-driven journey where you’ll:
Build a real JSON parser, in Rust 🏗️
Did you know about the following 5 Python 🐍 debugging tips? 🧵
January 21, 2026 at 2:40 PM
Did you know about the following 5 Python 🐍 debugging tips? 🧵