Published by
RDLB Agentic
Issue Nº 02
May 2026
A monthly field report from the agentic era

Brave
New Word

The production issue. Agents stopped demoing and started shipping work — coding the repo, running the tests — and ran straight into the reliability tax nobody had priced in.

Into ProductionThe Reliability TaxRequired Reading ×4Copilot's measured 55.8%
The Index

The month agents went to work.

01
From the editor's desk
Issue two. Capability was never the question. Reliability is.
03
02
Into production
Coding agents leave the chat window and start editing the repo, running the tests, opening the PR.
04
03
The reliability tax
The metric that separates a clever agent from a dependable one — and why most fail it.
06
04
Required Reading
Four papers on shipping and measuring agents, each with a citation, a link, and the line that matters.
07
05
Case study — Copilot's 55.8%
A randomized trial, not a testimonial. What measured productivity actually looks like.
08
06
The Ticker
Four real signals from agents in production, sourced and dated.
09
07
Notes from the floor
RDLB Agentic in practice — evals before prod, reliability as a gate, judgment kept human.
10
08
Sources & colophon
Every claim, linked. Plus the type, the tools, and next month.
11
Masthead
Brave New Word is published monthly by RDLB Agentic — the operating arm of RDLB Agency. We build brand systems that think, create, and operate. We do not sell AI tools.
Editorial standard
Every figure is concrete or absent. Every claim is sourced. One enemy per piece. Mechanism over metaphor. If we can't link it, we don't print it.
Brave New Word Issue Nº 02 · May 202602
01 · From the editor's desk

Issue two. Agents went to work.

Last month's question was "can it act?" This month's is harder and far more expensive: "can it act dependably, on the hundredth try, unattended?" The agentic economy does not turn on a model's best day. It turns on its worst one.

Production is where the romance ends and the engineering begins. A coding agent that resolves a GitHub issue in a demo is a marvel. A coding agent you can point at a backlog overnight is a business. The distance between those two sentences is this entire issue.

We walk through the systems that crossed it — agents with a real interface to a computer, editing files and running tests — and then we do the unglamorous thing and measure them. The headline result of the year is not a capability score. It is a reliability score, and it is sobering: even strong agents that succeed once often cannot succeed consistently.

The enemy this month is vibes-based autonomy — the decision to let an agent run unattended because a demo felt impressive, without a number for how often it actually finishes the job correctly. Feelings are not a deployment criterion. Pass rates are.

The encouraging half: when you do measure honestly, the value is real and large. A randomized trial — not a testimonial — is our case study. Read it, then go ask your vendors for their pass rates.

The agentic economy doesn't turn on a model's best day. It turns on its worst one.

— RDLB Agentic, editorial. New York. rdlbagentic.com

Brave New Word Issue Nº 02 · May 202603
02 · Production — Feature

Into production.

The clearest proving ground for working agents is software itself, because the work is verifiable: did the test pass or not? That is why the most important production benchmark of the era is about resolving real bugs in real codebases.

"SWE-bench" (Jimenez et al., 2023) set the bar: take real GitHub issues from popular Python projects and ask a model to produce a patch that actually resolves them, across multiple files.1 When it launched, the verdict was humbling — existing models could handle only the simplest tasks. The benchmark made "can it code?" precise: not "write a function," but "fix this repository."

The leap came from giving the model a real way to work. "SWE-agent" (Yang et al., 2024) introduced an agent-computer interface — a purpose-built set of commands to navigate a repo, edit files, and run tests — and showed that interface design, not just model size, drives results.2 With it, an agent reached a pass@1 of 12.5% on SWE-bench, far above the non-interactive baselines of the time. A small number, but a structural one: the agent was now operating the codebase, not describing it.

This is the production pattern in miniature. Localize the problem, edit, run the tests, read the failure, try again — the perceive-reason-act-reflect loop from last issue, pointed at a build system. When the environment can tell the agent whether it succeeded, the loop closes and real work gets done.

The catch is that "12.5% of the time" is also a sentence about the other 87.5%. Which is where reliability — the subject of the next page — stops being academic.

A coding agent that resolves an issue in a demo is a marvel. One you can point at a backlog overnight is a business.
Brave New Word Issue Nº 02 · May 202604
Into production — continued

The production loop, plainly.

01 · Localize

Find the right file

Read the issue, search the repository, narrow thousands of lines to the few that matter. Most failures start here — fix the wrong place and nothing downstream helps.1

02 · Edit

Make the change

Apply a patch through a real interface — open the file, edit the lines, save. SWE-agent's insight: the quality of that interface is itself a major lever on success.2

03 · Test

Let reality vote

Run the suite. The test result is the ground truth that closes the loop — the rare case where an agent gets unambiguous, automatic feedback on whether it actually worked.

04 · Iterate

Read the failure, retry

Parse the stack trace, revise, run again — until green or until the budget is spent. The discipline is knowing when to stop and escalate to a human.

Why a brand operator should care

The reason coding agents work better than most is that the environment grades them automatically. Your marketing work rarely does. So the production question for any non-code agent is: what is your test suite? If you cannot define an automatic check for "did it do the job right," you cannot safely let it run unattended — you can only let it draft.2

Brave New Word Issue Nº 02 · May 202605
03 · Innovation watch

The reliability tax.

Capability and reliability are different axes, and the gap between them is the single most under-priced cost in agentic AI. A 2024 benchmark from Sierra named it, and the number is the one to remember.

"τ-bench" (Yao et al., 2024) does something most benchmarks don't: it talks back. It simulates a real user conversing with a tool-using agent under domain rules — a retail or airline help desk — and checks the actual end state of the database, not just the agent's confident summary.3

Then it does the thing that matters for production: it runs each task many times. Its pass^k metric measures whether an agent can succeed not once but consistently. The finding is the headline of this issue — even state-of-the-art agents solved fewer than half the tasks on a single try, and when asked to repeat the same task eight times, consistency fell to roughly 25%.3 The capability was there. The dependability was not.

This is the reliability tax. An agent that works 80% of the time is not "mostly done" — at scale it is a steady stream of wrong actions you must catch, reverse, and apologize for. The metric is now serious enough that frontier model cards report pass^k directly.3,5

<50%
top agents, success
on a single try3
~25%
consistency over
8 repeats (pass^8)3
pass^k
the metric now in
frontier model cards5
Brave New Word Issue Nº 02 · May 202606
04 · Required Reading

Four papers on shipping.

From benchmark to interface to reliability to measured value. Each freely available and verified live for this issue.

Jimenez, Yang, Wettig … (2023) · arXiv:2310.06770

SWE-bench: Can Language Models Resolve Real-World GitHub Issues?

arxiv.org/abs/2310.06770

Why it mattersRedefined "can it code" as "can it fix this repository." The benchmark that turned coding agents from a demo into a measurable discipline.

Yang, Jimenez, Wettig, Lieret, Yao, Narasimhan, Press (2024) · arXiv:2405.15793

SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering

arxiv.org/abs/2405.15793

Why it mattersThe interface is the product. A well-designed agent-computer interface, not just a bigger model, is what let agents actually operate a codebase.

Yao, Shinn, Razavi, Narasimhan (2024) · arXiv:2406.12045

τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains

arxiv.org/abs/2406.12045

Why it mattersIntroduced pass^k and proved the reliability tax: strong agents that succeed once routinely fail to succeed consistently. The number every buyer should demand.

Peng, Kalliamvakou, Cihon, Demirer (2023) · arXiv:2302.06590

The Impact of AI on Developer Productivity: Evidence from GitHub Copilot

arxiv.org/abs/2302.06590

Why it mattersA randomized controlled trial, not a testimonial: the treatment group finished 55.8% faster. This is what measured value looks like.

Brave New Word Issue Nº 02 · May 202607
05 · Case study

Copilot's 55.8%, measured properly.

Almost every productivity claim in AI is a vibe with a logo. This one is a randomized controlled trial — which is exactly why it is worth more than a hundred case-study quotes.

55.8%
faster task completion,
treatment vs control4
3
settings: Microsoft,
Accenture, a Fortune 1004
RCT
randomized — not a
testimonial4

In 2023, researchers at Microsoft and GitHub ran a controlled experiment: developers were asked to build an HTTP server in JavaScript, randomly assigned to use GitHub Copilot or not.4 The group with the AI pair programmer finished 55.8% faster. The result was replicated across field experiments at Microsoft, Accenture, and a large anonymized firm.

Two details make it credible rather than hype. First, randomization: the only systematic difference between groups was the tool, so the speed-up is attributable to it, not to motivated volunteers. Second, the heterogeneity: less-experienced developers adopted it more and gained the most — a clue that these systems compress the distance between junior and senior on well-scoped work.

But read it next to τ-bench and the lesson sharpens. Copilot's win is a throughput result on a bounded task with a human in the loop at every keystroke. It is not a claim that an unattended agent finishes the job correctly 56% more often. Throughput scales with assistance; reliability is what you need before you remove the human.

Measured properly, the value is real and large. Measured loosely, it becomes the excuse for vibes-based autonomy. The difference is whether there was a control group.

The operator's takeaway

Demand two numbers, not one. Throughput (how much faster, with a human assisting) and reliability (pass^k: how often it finishes correctly, unattended). Copilot is a throughput story. Autonomy is a reliability story. Do not pay for one and deploy the other.

Brave New Word Issue Nº 02 · May 202608
06 · The Ticker — real-world signals

Agents that actually shipped.

Dated, sourced, and free of the word "10x."

Oct 2023

SWE-bench raises the bar. Resolving real GitHub issues becomes the standard test for coding agents — and the early scores make clear how far there is to go.1

2024 · NeurIPS

The interface becomes the product. SWE-agent shows a purpose-built agent-computer interface lifts performance sharply — design, not just scale, moves the number.2

Jun 2024

Reliability gets a number. Sierra's τ-bench introduces pass^k and the uncomfortable finding: consistency, not capability, is the bottleneck. Frontier model cards begin reporting it.3,5

2023→

Productivity, measured. The Copilot RCT puts a defensible number — 55.8% faster — on assisted developer work, and finds the largest gains among less-experienced developers.4

What we're watching next month

From shipping to speaking: how agents coordinate and communicate across systems — the protocol wars, and why the brand that picks well stops paying the integration tax. That is Issue Nº 03.

Brave New Word Issue Nº 02 · May 202609
07 · Notes from the floor — RDLB Agentic in practice

Evals before prod. Judgment stays human.

A production magazine should run on a production system. Here is how this issue's research disciplines our own build.

Every agent has a test suite. We borrow the coding-agent trick: define an automatic check for "did it do the job right" before anything runs unattended. No check, no autonomy — only drafting.2

We gate on pass^k, not pass^1. A capability that works once is a demo. We measure consistency across repeats and set a reliability bar an agent must clear before it touches a client's brand — because the worst day is the one that ships.3

Throughput to the agent, judgment to the human. Like the Copilot result, our agents compress the time-to-draft dramatically. They do not get the final say. A person signs the work, every time.4

What we refuse. Vibes-based autonomy. We will not let a system run unattended because a demo felt good. If we cannot show you the pass rate, we have not earned the right to remove the human — and we tell you so.

No automatic check for "did it do the job right" means no autonomy — only drafting.
Brave New Word Issue Nº 02 · May 202610
08 · Sources — every claim, linked

If we cited it, here it is.

01
Jimenez, C. E., Yang, J., Wettig, A., et al. (2023). SWE-bench: Can Language Models Resolve Real-World GitHub Issues? arXiv:2310.06770.
https://arxiv.org/abs/2310.06770
02
Yang, J., Jimenez, C. E., Wettig, A., Lieret, K., Yao, S., Narasimhan, K., Press, O. (2024). SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering. arXiv:2405.15793.
https://arxiv.org/abs/2405.15793
03
Yao, S., Shinn, N., Razavi, P., Narasimhan, K. (2024). τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains. arXiv:2406.12045.
https://arxiv.org/abs/2406.12045
04
Peng, S., Kalliamvakou, E., Cihon, P., Demirer, M. (2023). The Impact of AI on Developer Productivity: Evidence from GitHub Copilot. arXiv:2302.06590.
https://arxiv.org/abs/2302.06590
05
Sierra (2024). τ-bench: Benchmarking AI agents for the real world.
sierra.ai/blog/tau-bench-shaping-development-evaluation-agents
06
Microsoft Research. The Impact of AI on Developer Productivity: Evidence from GitHub Copilot (publication page).
microsoft.com/en-us/research/publication/the-impact-of-ai-on-developer-productivity…

Colophon
Set in three faces echoing the RDLB system: a geometric display sans, a humanist Garamond for body and its italic payload word, and JetBrains Mono for labels. This issue leads in constellation neon — the Production accent. Black, white, and the constellation palette only.
Next issue · No. 03
The protocol wars: how agents learn to talk to each other — MCP, A2A, ACP, ANP — and the integration tax you stop paying. Out July 2026.
Brave New Word Issue Nº 02 · May 202611
RDLB Agentic
Brave New Word · Nº 02
The single thing to do after reading this

Map your highest-cost work.

Book a 30-minute strategy blueprint call. You leave with a one-page map of your highest-cost work, sorted three ways — what to replace, what to augment, and what to leave alone — against real numbers, not vibes.

dashboardrdlbagency.com/book

Brand strategy, creative intelligence, and agentic systems — the operating arm of RDLB Agency. We build brand systems that think, create, and operate continuously.
© 2026 RDLB Agentic
Issue Nº 02 · the Production number