Weco AI
@WecoAI
Followers
1K
Following
63
Media
20
Statuses
37
The Platform for Self-Improving Code
Joined April 2023
Hard work scales linearly. Automation scales exponentially. Over 17 days, our autonomous ML agent trained 120 models and beat 90% of teams in a live $100k ML competition, with zero human intervention. Weco, now in public beta:
1
21
34
Early users are already using it for prompt engineering, kernel engineering, model development, agent development, and more. See more details: https://t.co/CeF7wpooM7 Or get started now: https://t.co/jCjbcXkar4 We’ll give the first 2,000 users $20 in credits each.
weco.ai
Hard work scales linearly. Automation scales exponentially.
0
0
4
Given only the competition description as input, Weco ranked 34th out of 490 teams in the CrunchDAO ADIA Structural Break Challenge, a $100k ML competition. Over 17 days, it explored 120 solutions completely autonomously. Check here for all of the experiments Weco has run:
dashboard.weco.ai
Weco Dashboard for viewing optimization runs
1
0
3
Weco, at its core, runs an evaluation-driven code optimization loop. Start with your repo and a metric. Weco keeps proposing, running, and measuring ideas until the curve bends upward. Every idea gets tested by experiment, not just your hunches.
1
0
4
Exciting things to come...👀
Thrilled to announce Weco has raised an $8M seed led by @GoldenVentures to build self-evolving software! Our technology has already been used by frontier labs like OpenAI, Meta, Google and Sakana AI. We’re making every codebase a living experiment that learns to beat itself:
0
1
14
🔗 Papers & code Speedrunning Benchmark → https://t.co/bATA0Fkp5R AIRA agents → https://t.co/ALBOIzW6jN AIDE ML, our reference implementation of AIDE (MIT) →
github.com
AIDE: AI-Driven Exploration in the Space of Code. The machine Learning engineering agent that automates AI R&D. - WecoAI/aideml
0
0
1
Thrilled to see @WecoAI's AIDE used in Meta's work and big congrats to @MinqiJiang @BingchenZhao @MarlaMagka. It's a truly exciting time to work on Recursive Self-Improvement.
Recently, there has been a lot of talk of LLM agents automating ML research itself. If Llama 5 can create Llama 6, then surely the singularity is just around the corner. How can we get a pulse check on whether current LLMs are capable of driving this kind of total
2
4
16
Solid work from @AIatMeta on ablating and improving AIDE on MLE-Bench! The rigor of empirical evaluation has reached a new level, making the experimental signals super strong. Highly recommended for anyone interested in AI-Driven R&D/Agentic Search!
AI Research Agents are becoming proficient at machine learning tasks, but how can we help them search the space of candidate solutions and codebases? Read our new paper looking at MLE-Bench: https://t.co/uX09L8zOBi
#LLM #Agents #MLEBench
0
7
42
To dive deeper, check out our paper: https://t.co/hWmwRYCW0U or explore the code on GitHub: https://t.co/lFEpKJbMwv Excited to see what the community builds with AIDE! 🚀
arxiv.org
Machine learning, the foundation of modern artificial intelligence, has driven innovations that have fundamentally transformed the world. Yet, behind advancements lies a complex and often tedious...
0
0
14
OpenAI's MLE-Bench shows that o1-preview + AIDE excels at ML engineering, but how much does AIDE contribute? Our tests confirm that AIDE boosts performance 3.5x over o1-preview alone. (4/N) https://t.co/hi7h0KhwBY
We’re releasing a new benchmark, MLE-bench, to measure how well AI agents perform at machine learning engineering. The benchmark consists of 75 machine learning engineering-related competitions sourced from Kaggle.
2
4
17
We’re sharing more details on our internal Kaggle benchmark. Unlike MLE-Bench, we made actual Kaggle submissions whenever possible. Here, we provide insights into the benchmark setup, key results, and the limitations of our evaluation protocol. (3/N)
1
0
7
We break down AIDE's algorithm to highlight its design philosophy that enables interaction scaling. AIDE uses a systematic tree search, iteratively refining solutions with improvements or bug fixes while evaluating performance at each step. (2/N)
1
0
7
AIDE has stood the test of time as the leading ML engineering agent, showing strong potential to automate data science modeling, deep learning, and AI R&D. Today, we’re sharing more details to help the community better understand its design, and build on top of it:🧵 (1/N)
7
29
130
🏃♀️ MLE-bench Lite 🪶 The most common request we get on MLE-bench is to have a “Lite” version that is cheaper to run, and we’re happy to share this today! TL;DR: No new datasets required, simply report results on the “Low” complexity split of existing MLE-bench comps.
2
9
76
Handwriting Calculator Typing math symbols is a pain- why not just draw them? Our AI can handle advanced math, from basic arithmetic to integrals. Try it here: https://t.co/RQ4jOVhhSF (3/N)
0
1
5
Easy Deployment & Spreadsheet Integration Use our single-line deploy approach or integrate Weco AI Functions directly in Google Sheets—batch-process spreadsheet data with ease. Get the add-on: https://t.co/6TJ2cf0XRW (2/N)
1
0
4
🚀 AIDE just got better! Our ML engineering agent now has a local Web UI: ✨ Visual interface for seamless ML experiments 📊 Better tracking & progress monitoring 🔒 Fully local—no data uploads required Try it now: https://t.co/CjxWDTMXMb
5
4
22
We’re looking for a frontend engineer to join us in building interfaces that deliver AI features with a prompt! Time zone preference: US East or UK Location: Flexibility to relocate to the Bay Area is preferred DM this account or @zhengyaojiang if you're interested
2
4
12
Excited to see OpenAI's recent project, MLE-bench, is based on our open-source effort, AIDE. In their independent evaluation, AIDE surpasses other MLE agents by a large margin!
We’re releasing a new benchmark, MLE-bench, to measure how well AI agents perform at machine learning engineering. The benchmark consists of 75 machine learning engineering-related competitions sourced from Kaggle.
10
24
139