Survey from
SF Ruby
Conf '26

The state of
agentic
engineering 2026

We surveyed engineers, tech leads, and founders at SF Ruby Conf about how they work with coding agents: which ones they use, whether they run them locally or in the cloud, and who, if anyone, reviews the code.

~/survey — zsh
$ sc survey --summary
82%say agents wrote most or all of the code they shipped
78%talk to their agents from a terminal
23%regularly prompt agents from their phone
61%want to work in shared agent sessions but don't yet
$
$ sc survey --who-answered

Who answered the survey

Just over half are engineers, and 57% work at companies of 50 people or fewer.

roles.tsv

Role

Engineer
53%
CTO / Founder
22%
Manager / Director
11%
Tech lead
10%
Product, growth, marketing, student
5%
company_size.tsv

Company size

Just me
13%
2–10
13%
11–50
31%
51–200
21%
201–1,000
11%
More than 1,000
12%
# The findings
$ sc survey --who-writes-the-code
82%

say agents wrote most or all of the code they shipped last month

Agents write most of the code now.

54% say agents wrote roughly all of the code they shipped in the past month, and 28% say most of it. Only 1% say none of it.

shipped_last_month.tsv

How much of the code you shipped in the past month was written by coding agents?

54%Roughly all of it
28%Most of it
17%Some
1%None
$ sc survey --agents
84%

use Claude Code every day

Nearly everyone uses Claude Code daily.

Everyone in the survey has tried it, and only 7% tried it and stopped. Codex is second, with 20% using it daily. For Cursor Agent and GitHub Copilot Agent, more respondents have stopped using them than use them now.

agents.tsv

Which coding agents have you tried?

Use dailyUse sometimesTried it, stoppedHaven't tried it
Claude Code
84%
10%
Codex
20%
27%
21%
32%
GitHub Copilot Agent
10%
20%
38%
32%
Cursor Agent
10%
12%
38%
41%
OpenCode
12%
78%
Pi
86%
Grok Build
89%
Amp
91%

69% use two or more agents, daily or sometimes.

$ sc survey --interface
78%

talk to their agents from a terminal

Most people still work from the terminal.

The terminal is well ahead of IDEs, which are at 41%. Agent development environments (ADEs) are much less common, at 12% local and 11% cloud.

interface.tsv

On a typical day, what is your interface to coding agents?

Terminal
78%
IDE (VS Code, Cursor, Zed)
41%
Local ADE (Conductor, Superset, cmux)
12%
Cloud ADE (Devin, Factory, Superconductor)
11%

34% use the terminal and nothing else.

$ sc survey --phone
23%

regularly prompt agents from their phone

Fewer than a quarter regularly prompt agents from their phone.

Another 38% do it sometimes, and 38% never do.

phone.tsv

How often do you chat with your coding agents from your phone?

Regularly
23%
Sometimes
38%
Never
38%
$ sc survey --spend
59%

pay $100 a month or more for coding agents

Most people pay $100 a month or more.

34% are on a $100–200 plan and 25% run more than one $200 plan. Together that is 59%. Another 28% are on a $20–50 plan, and 9% say their company pays. Respondents whose agents write nearly all of their code tend to be on the larger plans.

plans.tsv

Which coding agent plans do you use?

$100–200 / month
34%
More than one $200 / month plan
25%
$20–50 / month
28%
Company-paid: enterprise, team, or API billing
9%
No paid plan
3%
No idea who pays
2%
25%

run more than one $200 plan at the same time.

$ sc survey --review
78%

have an agent review code before a person does

Agents review first. A person still approves.

About four in five say a person must approve a change, and for 46% it has to be someone other than the person who prompted the agent. 14% merge with no human review once CI and agent reviews pass.

review_process.tsv

What is the standard code review process for features and bugfixes?

Someone other than the person who prompted the agent must review
46%
A person must review, and it can be the person who prompted
33%
No human review: merge once CI and agent reviews pass
14%
Depends on the change, often a bot approving low-risk PRs
7%

At companies over a thousand people, 83% need a second person. Among solo developers, 43% merge with no human review at all.

review_agents.tsv

What is your agentic code review setup?

One code review agent
53%
Several code review agents
25%
No agentic code review
22%

A quarter use several code review agents.

$ sc survey --non-engineers
41%

say non-engineers change the product with coding agents

Non-engineers are changing the product.

Most commonly (36%), non-engineers prompt coding agents and engineers review the result. 23% say they ship changes themselves, like landing pages or small fixes. About half say non-engineers use agents only for non-code work.

non_engineers.tsv

How do non-engineers on your team use agents?

They prompt coding agents to change the product, and engineers review it
36%
They build internal apps for themselves
31%
They ship changes themselves, like landing pages or small fixes
23%
They use agents, but only for non-code work
51%
Not sure
10%
They don't use agents
9%
slack.tsv

Does your team use AI agents inside Slack or Teams?

Yes, for code and non-code work
23%
Yes, for code work
9%
Yes, for non-code work
7%
No
62%

38% use agents in Slack or Teams. At companies with more than two hundred people, it's 70%.

$ sc survey --multiplayer

Teams want shared agent sessions.
Few have them.

We asked if two or more people on the team had ever worked with the same agent in the same conversation.

61%

haven't shared an agent session, but want to

11%

share agent sessions regularly

11%Yes, regularly
10%We've tried it
61%No, but we'd like to
19%No, and we don't need to

Excluding solo developers, 63% of people on a team want to work in shared agent sessions but don’t yet. One person's answer for how they share an agent: "Screen sharing…"

What a shared
agent session looks like

Work with your team in shared agent sessions.

Shared context from ideation, through implementation, and all the way to code review and merge.

Try shared agent sessions
Stellar › web › #482 Search bar in the hero
#1Claude Code · Opus 5.5
Submit PR
Arjun Singh9:12 AM
Visitors can't find trips from the home page. Add a search bar to the hero with destination and dates. @Olga FYI
Claude Code9:31 AM
Done: a SearchBar in the hero that filters trips by city and week.
6 files changed24 tests passingPreview updated
Olga Stadie9:40 AM
@Ibrahim can you check it on mobile first?
Ibrahim Awwal9:44 AM
Works on mobile. Claude, make the Search button green to match the brand.
Claude Code9:46 AM
Done. The button now uses the brand green, and the preview is updated.
Follow up with Claude Code…
Preview/
StellarTripsStaysSign in
Find your next escape
Trips and stays, planned with the people you travel with.
Kyotofrom $890
Lisbonfrom $640
Reykjavíkfrom $1,120
Oaxacafrom $520
Full DiffGuided ReviewQA 6/6
Step 2 of 4
Filter trips in SQL, not in Ruby
The form's params go to a Trip.search scope, so trips#index stays one line and Postgres does the filtering.
app/models/trip.rb+6
14scope :search, ->(city: nil, week: nil) do
15 rel = all
16 rel = rel.where(city:) if city.present?
17 rel = rel.where(departs_on: week) if week
18 rel
19end
week is a Range of dates, so line 17 runs as a single BETWEEN query.
db/migrate/…_add_search_index.rb+5
3add_index :trips, [:city, :departs_on]

Multiplayer Agent Chat

Build together in shared agent chats. Anyone on the team can jump in.

Live App Preview

Every agent runs your app on its own cloud computer. No local checkout.

Code Review

Read the diff with QA results and agent-guided review.

$ sc survey --add-your-answers

How does your team compare?

The survey is still open. Answer the same questions and see where your team sits next to these numbers.

Take the survey

# Superconductor ran this survey at SF Ruby Conference in San Francisco, November 10–12, 2026. Answers are self-reported and come from conference attendees, a crowd that leans toward Ruby, startups, and early adoption of coding agents. Results are aggregated; no individual response is identifiable. Percentages are rounded to the nearest whole point, so some charts sum to a point over or under a hundred. "Pick all that apply" questions sum to more.

Stop
copy-pasting
your agent output.

Share agent sessions and context from ideation, through implementation, and all the way to code review and merge. Bring your Claude Code, Codex, or other subscriptions.

$ sc demo --with-founders Three months of the Pro plan, free, when you schedule a demo with our founders.