Paul Twist

European AI engineer • Technical writer • I turn messy AI infrastructure into practical guides developers can actually use.

Berlin, Germany Joined on Jun 11, 2026

Education

Computer Science, distributed systems, and open-source infrastructure

Pronouns

he/him

Work

AI Engineer & Technical Writer

Paul Twist

Jun 29

The Skill Portability Problem: Why Your Best Agents Are Locked Into One Runtime

#ai #agents #infrastructure #litellm

5 min read

Paul Twist

Jun 28

The Four-Layer Agent Stack: When Your Framework Isn't Enough

#ai #agents #infrastructure #litellm

6 min read

Paul Twist

Jun 27

LiteLLM-Rust Changes Agent Memory Architecture: A 150x Speedup Shifts the Economics

#litellm #agents #infrastructure #rust

3 min read

Paul Twist

Jun 26

What Your Production Agents Aren't Telling You: A Practical Guide to Agent Observability

#ai #agents #observability #infrastructure

4 min read

Paul Twist

Jun 26

The Brain/Sandbox Pattern: Why Your Production Agent Needs This Architecture

#agents #architecture #production #infrastructure

7 min read

Paul Twist

Jun 26

Oracle Cloud Just Made LiteLLM a Native Provider for OCI Generative AI

#discuss #ai #cloud #programming

2 min read

Paul Twist

Jun 24

Your Agents Need a Security Boundary. Heres Why Its Become Non-Negotiable.

#ai #infrastructure #security #agents

5 min read

Paul Twist

Jun 23

LiteLLM vs Bifrost: I Tested Both in Production. Here's What Actually Matters.

#discuss #ai #webdev #programming

4 min read

Paul Twist

Jun 23

The Agents That Actually Ship: Why Boring Beats Autonomous

#ai #agents #infrastructure #production

4 min read

Paul Twist

Jun 23

LiteLLM Is Moving to Rust. Here's What the Benchmarks Look Like.

#discuss #ai #rust #programming

2 min read

Paul Twist

Jun 22

The Orchestration Bottleneck: Why Your Agent Infrastructure Needs Two Layers in 2026

#agents #infrastructure #orchestration #llm

5 min read

Paul Twist

Jun 21

Agent Session Memory Isn't a Feature. It's Your Control Plane.

#ai #agents #infrastructure #devops

6 min read

Paul Twist

Jun 20

We Cut Our LLM API Bill 30% With Four Lines of YAML

#discuss #ai #devops #programming

3 min read

Paul Twist

Jun 20

Why Your Agents Need Intelligent MCP Routing (And Why It's Harder Than It Looks)

#agents #mcp #infrastructure #routing

6 min read

Paul Twist

Jun 19

We Let 40 Engineers Loose With Coding Agents. The Bill Was Brutal.

#discuss #ai #llm #programming

3 min read

Paul Twist

Jun 19

The Multi-Runtime Agent Problem: Why Your Team Needs More Than One Runtime

#agents #infrastructure #llm #litellm

5 min read

Paul Twist

Jun 19

AWS Just Made LiteLLM a First-Class Model Provider in Amazon Bedrock AgentCore

#aws #ai #llm #programming

3 min read

Paul Twist

Jun 18

Stop Measuring Agent Infrastructure by Gateway Latency Alone

#ai #agents #infrastructure #performance

6 min read

Paul Twist

Jun 18

We Obsessed Over Gateway Latency for a Month. Then We Looked at the Actual Numbers.

#discuss #ai #llm #infrastructure

4 min read

Paul Twist

Jun 18

Why Your Agents Are Silently Burning Tokens (And How to Stop Them)

#ai #agents #infrastructure #production

5 min read

Paul Twist

Jun 17

Why Coding Agents Need Two Halves of Infrastructure: Control Plane + Fast Data Plane

#agents #llm #infrastructure #devops

7 min read

Paul Twist

Jun 11

The Missing Operational Layer Between Agent Prototypes and Production

#agents #ai #architecture #devops

4 min read

DEV Community

Paul Twist

Badges

Writing Debut

Skills/Languages

Currently learning

Currently hacking on

Available for

The Skill Portability Problem: Why Your Best Agents Are Locked Into One Runtime

The Four-Layer Agent Stack: When Your Framework Isn't Enough

LiteLLM-Rust Changes Agent Memory Architecture: A 150x Speedup Shifts the Economics

What Your Production Agents Aren't Telling You: A Practical Guide to Agent Observability

The Brain/Sandbox Pattern: Why Your Production Agent Needs This Architecture

Oracle Cloud Just Made LiteLLM a Native Provider for OCI Generative AI

Your Agents Need a Security Boundary. Heres Why Its Become Non-Negotiable.

LiteLLM vs Bifrost: I Tested Both in Production. Here's What Actually Matters.

The Agents That Actually Ship: Why Boring Beats Autonomous

LiteLLM Is Moving to Rust. Here's What the Benchmarks Look Like.

The Orchestration Bottleneck: Why Your Agent Infrastructure Needs Two Layers in 2026

Agent Session Memory Isn't a Feature. It's Your Control Plane.

We Cut Our LLM API Bill 30% With Four Lines of YAML

Why Your Agents Need Intelligent MCP Routing (And Why It's Harder Than It Looks)

We Let 40 Engineers Loose With Coding Agents. The Bill Was Brutal.

The Multi-Runtime Agent Problem: Why Your Team Needs More Than One Runtime

AWS Just Made LiteLLM a First-Class Model Provider in Amazon Bedrock AgentCore

Stop Measuring Agent Infrastructure by Gateway Latency Alone

We Obsessed Over Gateway Latency for a Month. Then We Looked at the Actual Numbers.

Why Your Agents Are Silently Burning Tokens (And How to Stop Them)

Why Coding Agents Need Two Halves of Infrastructure: Control Plane + Fast Data Plane

The Missing Operational Layer Between Agent Prototypes and Production