Tyler Edwards

Ask me about agents! Previously spent 7 years as a Software Engineer at MI5 & MI6 (GCHQ). Unfortunately I've been sworn to secrecy so can't answer any questions about that.

London Joined on Aug 21, 2026 https://www.overmindlab.ai

Tyler Edwards

Sep 30

Secret Agent Overmind: diagnose an AI agent with code, traces, evals and datasets

#ai #llm #observability #agents

7 min read

Want to connect with Tyler Edwards?

Create an account to connect with Tyler Edwards. You can also sign in below to proceed if you already have an account.

Create Account

Already have an account? Sign in

Tyler Edwards

Sep 29

Unwrapping the wrappers: who actually trains their own models, and why

#ai #llm #machinelearning #finetuning

7 min read

Tyler Edwards

Sep 23

The anatomy of an AI agent: model, loop, tools, memory

#ai #llm #agents #architecture

6 min read

Tyler Edwards

Sep 22

Think smaller: why specialist SLMs beat frontier models in production

#ai #llm #machinelearning #opensource

8 min read

Tyler Edwards

Sep 8

Prompt engineering vs fine-tuning: which one do you need?

#llm #promptengineering #finetuning #ai

4 min read

Tyler Edwards

Sep 7

Open-weights LLMs vs frontier APIs: when to rent, when to own

#llm #opensource #ai #machinelearning

5 min read

Tyler Edwards

Sep 4

How do you turn traces into a training dataset?

#llm #machinelearning #ai #observability

7 min read

Tyler Edwards

Sep 1

What are the different types of fine-tuning?

#llm #finetuning #machinelearning #ai

6 min read

Tyler Edwards

Sep 1

"So, you have observability. Now what?"

#ai #llm #observability #machinelearning

9 min read

DEV Community

Tyler Edwards

Badges

Writing Debut

Available for

Secret Agent Overmind: diagnose an AI agent with code, traces, evals and datasets

Want to connect with Tyler Edwards?

Unwrapping the wrappers: who actually trains their own models, and why

The anatomy of an AI agent: model, loop, tools, memory

Think smaller: why specialist SLMs beat frontier models in production

Prompt engineering vs fine-tuning: which one do you need?

Open-weights LLMs vs frontier APIs: when to rent, when to own

How do you turn traces into a training dataset?

What are the different types of fine-tuning?

"So, you have observability. Now what?"