Jeffrey 🐬 confident-ai.com
@jeffr_yyy
Followers
159
Following
282
Statuses
898
Cofounder @confident_ai, building @deepeval, ex-@Google, ex-@Microsoft
San Francisco
Joined September 2020
It's official! @confident_ai just went public on @ycombinator's Launch YC! Checkout how we're changing LLM evaluation, our open-source approach, and ROIs we're bringing for customers:
0
8
17
0
0
1
The o-1 models overthink too much at times, not necessarily better than got-4o, just different
0
0
1
RT @KritinV78426: LLM evaluation has come a long way—AI agent evaluation hasn’t. Most existing benchmarks (AgentBench, SWE-bench) focus on…
0
2
0
RT @ycombinator: Subtrace lets developers resolve production issues faster. Unlike Sentry and other backend monitoring tools, @subtrace_de…
0
8
0
RT @jeffr_yyy: It's official! @confident_ai just went public on @ycombinator's Launch YC! Checkout how we're changing LLM evaluation, our…
0
8
0
Uploading our launch yc video to the exact page on youtube creators studio i've built back when i was at google is an interesting experience
0
0
2
from personal experience, research-based OS projects cannot live solely off of their research paper
0
0
0
@simon_kubica @blackbirdvc @BainCapVC @indexplan congratz! hows this different from clickup? or is this just a linear killer?
0
0
0
And @deepeval was used for evals btw
NEW § LLMs struggle with large amounts of context. Bharani Subramaniam and I explain how to mitigate this common RAG problem with a Reranker which takes the document fragments from the retriever, and ranks them according to their usefulness.
0
0
2
As startup founders, how do you think about competition and price wars? Is there a tendency to race to the bottom?
0
0
1
I just found this gemstone of an article written by @martinfowler. He shows how LLM applications (mainly RAG) are built and even use @deepeval in its evaluation code samples! Check if out here:
0
0
2
You know ChatGPT is about to cook when its response starts with “Unforunately”
0
0
0