Benchmark DashboardConfident AI
AI Infrastructure · 3.6K followers · LinkedIn benchmark, last 30 days (original)
Momentum vs prior 30 days
Posts
13
+550%Posts/wk
3
+500%Avg engagement
19.1
smart avg / post
-65%Momentum
not enough posts
2nd half of window vs 1st
Engagement trend
Posting volume (bars) vs average engagement per post (line) — page + team, original posts
Momentum over time
Change in average engagement per post vs the previous week — green gaining, red losing
Category mix
Cadence
Average engagement by day posted (Pacific time)
Team contribution
Team data through 2026-08-12 — employee scrapes lag this window; low numbers here mean stale data, not a quiet team.
What drives engagement
Smart avg engagement per post across every post in this window. Gray rows at the bottom are under the sample floor (8+ posts per category, 8+ per format) — too few posts to trust the average.
By category
By format
Timing detail is in the Cadence chart above.
Best performing posts
- 2026-08-07 · Jeffrey Ip · Product Update49 eng. (38 likes · 7 comments · 4 shares)
Are you building AI agents using Typescript? A fan of frameworks such as Vercel's AI SDK, Mastra? Want to do evals locally, without a dedica…
View post → - 2026-08-12 · Jeffrey Ip · Product Update36 eng. (27 likes · 5 comments · 4 shares)
From Pytest for LLMs to Vitest for LLMs, today, I'm proud to announce that DeepEval for TypeScript is now fully open-source. TypeScript supp…
View post → - 2026-07-28 · Jeffrey Ip · Product Update36 eng. (28 likes · 3 comments · 5 shares)
Today we're releasing DeepEval v4.1.4, where we're allowing for users to mark test cases and metrics as flaky ❄️. We're seeing things move m…
View post → - 2026-08-03 · Jeffrey Ip · Product Update27 eng. (22 likes · 2 comments · 3 shares)
Today I'm happy to announce that Confident AI is officially live on Amazon Web Services (AWS) Marketplace! This has been one of our most req…
View post → - 2026-08-11 · Jeffrey Ip · Product Update25 eng. (22 likes · 1 comments · 2 shares)
The modern AI architecture is incredibly fragmented, which is why evals feel so difficult to understand. AI apps can be classified as single…
View post →
Worst performing posts — settled posts only (published ≥7 days before the data edge)
- 2026-07-30 · Kritin Vongthongsri · Product Update5 eng. (4 likes · 1 comments · 0 shares)
In real AI systems, evals are not always perfectly stable. DeepEval now lets teams mark specific test cases or metrics as flaky, so noisy fa…
View post → - 2026-07-30 · Jeffrey Ip · Product Update7 eng. (7 likes · 0 comments · 0 shares)
Some eval failures are real regressions. Some are just flaky. DeepEval now lets you mark the flaky ones so your CI can keep moving, while th…
View post → - 2026-08-04 · Company page · Product Update12 eng. (10 likes · 1 comments · 1 shares)
Agentic RAG can still make things up. Even when your app gives the model good context, the model can add one extra claim that was never in t…
View post → - 2026-07-30 · Company page · Product Update17 eng. (11 likes · 1 comments · 5 shares)
DeepEval now lets you mark noisy evals as flaky, without hiding their results. Test cases and metrics both support flaky=True. For test case…
View post → - 2026-07-29 · Jeffrey Ip · Product Update23 eng. (21 likes · 1 comments · 1 shares)
One toxic trait Confident AI - we build everything ourselves, including the docs website, and today, we're launching it in public. Any sane …
View post →
All posts in this window
Posts
0 posts found