Home
Podcast
Notes
Playbooks
Archive
About
Home
Podcast
Notes
Playbooks
Archive
About
Subscribe
Sign in
Inside the harness (Part 2): Give it a plan
How planning and nudges keep a ReAct loop from losing the plot
Sep 4
•
Tanya D'cruz
and
Henry Li
4
3
Inside the harness (Part 1): Start with the loop
How a weather lookup becomes deep research without changing the ReAct machinery underneath
Aug 28
•
Tanya D'cruz
and
Henry Li
9
1
2
52:00
Inside the realities of AI engineering
A roundtable with AI Engineering author Chip Huyen and LLM Engineer’s Handbook co-authors Paul Iusztin and Maxime Labonne
May 28
•
Tanya D'cruz
10
2
49:57
Free webinar: Building AI agent harnesses for finance
Nicole Königstein on why reliable financial agents are an engineering problem, not simply a model problem.
Aug 22
•
Tanya D'cruz
13
2
3
Lessons from the frontlines of small-model development
Why small models are harder than they look, with Maxime Labonne
Jun 10
•
Tanya D'cruz
8
Recent posts
View all
45:03
🎙️ Episode 3: The agent autonomy trap
Podcast with Ben Auffarth on bounded autonomy, calibrated confidence, meaningful human oversight and why businesses should fund the dataset before the…
Aug 20
•
Tanya D'cruz
and
Ben Auffarth
8
3
#8: You can't trust what you don't evaluate
NVIDIA's Balamurugan Balakreshnan maps the evaluation practices that separate reliable AI agents from convincing demos
Aug 13
•
Tanya D'cruz
and
Balamurugan Balakreshnan
15
3
47:57
🎙️ Episode 2: An agent loop needs a way out
Podcast with Rick Hightower on verification, stopping conditions, and the harnesses that keep autonomous agents honest
Aug 6
•
Tanya D'cruz
and
Rick Hightower
18
2
6
#7: The undeclared dependency
Every team building with AI coding agents is carrying a dependency it never declared. Here's what happens when you finally do.
Jul 30
•
Tanya D'cruz
and
Maxim Salnikov
9
1
4
59:31
🎙️ Episode 1: GraphRAG as an interface to the world
Podcast with David Knickerbocker on teaching GraphRAG agents when to stop, what to remember and how to investigate
Jul 23
•
Tanya D'cruz
and
David Knickerbocker
23
1
9
#6: From prompt to harness to loop
Mona Mona on the agentic skill ladder no one told you about
Jul 16
•
Tanya D'cruz
and
Mona Mona
17
2
2
#5: The three disciplines of agentic engineering
Ken Huang, founder of Agentic AI and author of OpenClaw AI in Production, maps the three disciplines shaping the future of production AI systems
Jul 9
•
Tanya D'cruz
51
3
9
#4: The illusion of autonomous agents (part 2)
Denis Rothman on context engineering as the missing layer for reliable AI agents
Jul 2
•
Tanya D'cruz
14
2
SUSE refuses to measure its engineers by how much code their agents write
Rick Spencer on why output, tokens, and lines of code tell you nothing, and what an open-source enterprise tracks instead
Jun 30
•
Tanya D'cruz
8
2
See all
Agentic Engineering
Where the hype ends, and the build begins
Subscribe
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
