EdotEnv (@edotenv) profile picture

Quant Neolab building RSI by using market as self-improving systems

SF
Joined August 2026
This is us in Market Reasoning Models! We use the market as self improving benchmark to train better reasoning & research models.
New @cbventures market map alert! 🚨 AI agents are moving from talking about markets to trading in them. But as agents get easier to build, where will lasting value accrue? My new post maps the agentic trading stack through that lens. Here’s where I’m focused: • Agentic trading harnesses and terminals that turn a user’s goal into an ongoing strategy, with real retention. • Market reasoning models and data assets that give trading agents a real edge. • Identity, audit trails, and performance verification infra built for non-human, financial decision makers. I care less about shiny demos and more about products that get harder to copy as they grow. If you’re building here, I’d love to hear from you.
1
3
511
EdotEnv retweeted
if you trust the AI 2027 timeline, then next year is the year of AI doing AI research to be ahead of the curve, we already now build autoresearch/ RSI envs on real world ML research tasks
2
1
49
10,679
Training for RSI with true OOS data and real world ML research tasks
We're introducing TasteBench, a new standard to measure Research Taste on real world ML tasks. Existing benchmarks have limited datasets and eval data ultimately becomes part of training data. In TasteBench, we continuously generate new ML tasks on OOS data no model has trained on.
4
379
we tried coding harnesses for research and had to reimplement many obvious research frameworks - so we ended up building a harness from scratch Introducing oh-my-quant (OMQ): a research-first harness that handles all the boring parts so the Agent focus on intelligent tasks OMQ automatically: - spawns agent-colleagues to double check work - sets up independent workspaces per experiment - monitors & hooks up experiments to any dashboard - proactively notifies you over telegram on progress - owns independent reviewers to prevent doom loops
4
1
2
35
8,328
here we use OMQ for autoresearch on predicting crypto volatility, but it goes without saying that is it not just restricted to crypto research but easily transferable to other domains
1
1
6
1,010
credits to our friends at rekursive AI for sagent and trackinizer! @hyperparticle
1
1
4
628
there is still a long way to go before LLMs are useful for genuinely hard research tasks we let GPT autoresearch Deep Learning alphas on crypto perp data and find that GPT: - doesn't explore parameters well, only minimal changes across iterations - has poor research taste, proposes experiments like three almost identical learning rates (wtf) struggles with intelligent evals, ie judging training curves which has no binary criteria
3
1
2
18
5,407
it never really explores parameter space extensively, but rather does minimal changes. here the cosine similarities of the hyper parameters across iterations (1=identical)
1
3
122
after discovering once that decreasing learning rate improved performance, it suggested three almost identical lrs for the next iteration ....
1
5
114
it never really finds new research directions, but rather tends to do minimal changes. here the cosine similarities of the hyper parameters across iterations
1
12
after discovering once that decreasing learning rate improved performance, it suggested three almost identical lrs for the next iteration ....
8
markets are the best dynamic benchmark for LLM training - no saturation - infinite data source - naturally noisy with distribution shifts
1
11
4,337
our harness + Opus on MLE bench lite also outperforms other harnesses MLE bench is the closest public benchmark to our research tasks, and we chose MLE lite because full MLE bench runs just take too long is the long runtime of ML experiments the reason why there are so few such benchmarks?
1
8
2,927
our general research harness defines a new Pareto frontier on terminal bench 2.1 improving general research skills seems to help with SWE tasks too here is how our harness works: - async orchestration layer - extremely parallelised execution runtime - dedicated communication system between main/ subagents - native research support: efficient tracking, identification & querying of experiments
1
6
1,053
we had the great honour to meet the OG YC agent @paulg lmao first 5 minutes he was like: I don't understand wtf you guys are doing
1
9
4,319
yesterday we hosted an incredible hack for RSI & autoresearch in SF with Autolab (@ottogin1 and the group are absolute pros at organising hackathons) we had almost 400 registrations and sadly could not take in everyone, but stay tuned for future hacks. teaser: there might be one just around the corner next Sunday on #RSI & Harnesses
1
1
2
8
2,758
and the beautiful office was provided to us kindly by fellow YC company usebourne.com - they've just raised $45m to build autonomous white collar workers
3
302