Working Surfacetopics

The ideas that keep coming up

13 topics gather the links that return to the same question. Pick one to see what it means and every piece filed under it.

Topic · 14 links · since 17 September 2026

Evals before the build

Before an AI feature is built, the team writes down what a good output looks like, so that the result can be tested against it.

Amazon Web Services
From prototype to production: Engineering agents with Agent DLC
Yuan Wei, Tianqin Cai and Jiade Wu · 12 November 2026
Three Amazon Web Services authors propose that AI agents be released only against written criteria a business owner approves, and show one prompt change improving one answer while worsening another.
How-to · agents · decisions
15 Nov 2026
Thoughtworks
SpecBench: A tool to evaluate AI-enabled software development workflows
Yixiong Jiang, Chen Chen and Yuhua Li · 13 November 2026
Thoughtworks engineers built SpecBench to measure AI coding workflows and found that adding specifications and planning raised correctness slightly while nearly doubling the time taken.
Company story · agents · decisions
14 Nov 2026
Nielsen Norman Group
Research-In-The-Loop: How UXR Adapts to Development at AI Speed
Tony Alicea · 13 November 2026
Tony Alicea of Nielsen Norman Group proposes that researchers add a research section to each feature's spec, so coding agents build the measurement along with the feature.
How-to · artifacts · collaboration
14 Nov 2026
Pinterest Engineering Blog
How Pinterest Improved Analytics Agent Compliance with Loop Engineering
Skyler Zheng, Chris Moradi, Lu Liu, Aman Gairola and Michael Chau · 12 November 2026
Pinterest engineers stopped hand-editing their data agent's instructions and built an automated loop that keeps only changes passing paired tests drawn from real chats.
How-to · agents · critique & review
13 Nov 2026
Microsoft Command Line
Nine engineers, 1,211 pull requests in a day: Lessons from building Copilot Cowork
Bas Brekelmans · 2 November 2026
Microsoft's Copilot Cowork team, where AI agents wrote most of the code, found that automated checks can pass without checking anything, and now makes every check prove it can fail.
Company story · critique & review · agents
5 Nov 2026
Microsoft Inside Track
Deploying AI with employee councils: An IT playbook from Microsoft
Jason Kellington · 29 October 2026
Microsoft's internal IT organisation replaced scattered AI experiments with six employee councils that choose projects, fix processes before automating them, review responsible use and measure results.
Company story · decisions · rituals
30 Oct 2026
Product Talk
Building an AI Voice Agent for 360 Degree Feedback: How Your360 AI Uses State Machines, Evals, and Layered Synthesis
Teresa Torres with Jared Goralnick, Eda Goksel Jackson and Alex Krasun · 15 October 2026
Your360 AI replaced a single prompt with a state machine, so its voice AI coach keeps a structured feedback conversation on track while still going deep.
How-to · agents · decisions
16 Oct 2026
BLOG@CACM
Nobody Did TDD for 25 Years. Now the Machine Requires It
Abtin Aghagolian · 28 September 2026
Abtin Aghagolian, chief technology officer of Pikd, a London augmented-reality company, argues that tests written before an AI agent writes code are now how engineers control its output.
How-to · agents · critique & review
28 Sep 2026
The Pragmatic Engineer
Design Engineering with Maggie Appleton
Gergely Orosz with Maggie Appleton · 23 September 2026
Maggie Appleton, a research engineer at GitHub, no longer reads the code AI agents write for her prototypes and writes a spec saying how the agent will check its work.
Company story · prototyping
26 Sep 2026
Anthropic
How to prepare for AI-driven code modernization projects
Jonah Ezekiel and Lexie Tonelli · 23 September 2026
Two Anthropic engineers advise companies modernising old code with AI agents to agree, before the work starts, what each change must prove and how much human review it gets.
Company story · critique & review · artifacts
24 Sep 2026
Lenny's Newsletter
Advanced evals: How to find (and fix) hidden AI failures in your product
Hamel Husain and Shreya Shankar · 22 September 2026
Two specialists in testing AI products argue that people should read real AI sessions and note the failures themselves before an agent looks for errors or anyone writes a metric.
How-to · critique & review · rituals
23 Sep 2026
Warp
Using LLM-as-a-judge scoring to measure your software factory
Zach Lloyd · 18 September 2026
Warp's chief executive explains how AI agents can grade a sample of coding agents' past work against written criteria, so that people review the failures rather than every change.
How-to · critique & review · artifacts
22 Sep 2026
The Pragmatic Engineer
AI Skills with Matt Pocock
Gergely Orosz with Matt Pocock · 17 September 2026
Engineer and educator Matt Pocock uses short instruction files that make AI coding agents question him closely about a plan, which leaves him more time for strategic work.
Company story · agents · decisions
20 Sep 2026
Product Talk
4 New Evals and 16 Experiment Variants to Fix 1 Customer Complaint
Teresa Torres · 16 September 2026
Teresa Torres spent three weeks and sixteen experiments fixing one customer complaint about an AI product, starting by measuring how often the error occurred.
How-to · critique & review
17 Sep 2026