Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile
Illya Gerasymchuk (httpz)

@illyagera

Crypto, Finance, Blockchain, AI, DeFi, Web3 and Zero-Knowledge. Pursuing master's in Law & Financial Markets. Building @zkLocus on @MinaProtocol | ☦️

ID: 581321028

linkhttps://illya.sh calendar_today15-05-2012 20:53:16

1,1K Tweet

437 Followers

157 Following

Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

The model factory has been refactored to use a DSL for configuring the pipeline -- so now you can visualize and configure the whole flow from a single place This also makes it easier for agents to autonomously discover and verify profitable trading strategies, but I'll cover

The model factory has been refactored to use a DSL for configuring the pipeline -- so now you can visualize and configure the whole flow from a single place

This also makes it easier for agents to autonomously discover and verify profitable trading strategies, but I'll cover
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

I've been running live a tuned model designed to predict Bitcoin's price moves in Polymarket's BTC's 5 minute up or down markets Backtests had a quite remarkable perfomance, with PnL over 70% across >500 trades As expected, real world performance is not as impressive, but it's

I've been running live a tuned model designed to predict Bitcoin's price moves in Polymarket's BTC's 5 minute up or down markets

Backtests had a quite remarkable perfomance, with PnL over 70% across >500 trades

As expected, real world performance is not as impressive, but it's
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

Using Neural Networks To Find Imbalances In Limit Order Books I've been working on model factory that predicts short-term asset movements. The model focuses on predicting whether the price of an asset at a future time T+c will be below or above the current asset's price at time

Using Neural Networks To Find Imbalances In Limit Order Books

I've been working on model factory that predicts short-term asset movements. The model focuses on predicting whether the price of an asset at a future time T+c will be below or above the current asset's price at time
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

So I extended my MCP server with a tool to execute arbitrary Python code, and now that's pretty much the only tool the host agent is using πŸ˜‚ Instead of reusing the other tools, the agent prefers to write the whole logic from scratch and then pass to the MCP to execute it. It

So I extended my MCP server with a tool to execute arbitrary Python code, and now that's pretty much the only tool the host agent is using πŸ˜‚

Instead of reusing the other tools, the agent prefers to write the whole logic from scratch and then pass to the MCP to execute it. It
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

Crude Oil Prices Continue To Increase Over the past year, I've reiterated multiple times on the fact that that oil prices will continue to increase and that they'll soon be back up above their June 2025 prices. This happened again today, with Brent Crude Oil being up more than

Crude Oil Prices Continue To Increase

Over the past year, I've reiterated multiple times on the fact that that oil prices will continue to increase and that they'll soon be back up above their June 2025 prices. This happened again today, with Brent Crude Oil being up more than
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

Why High Oil Prices Are Bad For USD (HINT: China & Credit) I don’t think that higher oil prices will help the value of US dollar medium to long-term this time. Higher oil prices induce higher prices for producers, which lead to higher prices for consumers. The US economy is

Why High Oil Prices Are Bad For USD (HINT: China & Credit)

I don’t think that higher oil prices will help the value of US dollar medium to long-term this time. Higher oil prices induce higher prices for producers, which lead to higher prices for consumers. The US economy is
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

I've been calling for new local heights for oil prices for over 6 months now Last time Brent Oil traded at $85 per barrel was in 2024 Throughout 2025 and start of 2026, some of the largest finance accounts on X where telling you to short oil. Some are still referring to these

I've been calling for new local heights for oil prices for over 6 months now

Last time Brent Oil traded at $85 per barrel was in 2024

Throughout 2025 and start of 2026, some of the largest finance accounts on X where telling you to short oil. Some are still referring to these
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

CUSUM Filter Visualized I’m looking into CUSUM Filters in more depth and just wanted to share this nice visualization that Gemini created to demonstrate graphically how the threshold affects the sampling/event emission for a price series. In this example, price is used a the

CUSUM Filter Visualized

I’m looking into CUSUM Filters in more depth and just wanted to share this nice visualization that Gemini created to demonstrate graphically how the threshold affects the sampling/event emission for a price series.

In this example, price is used a the
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

Very excited for the upcoming price rebound in commodity metals Current prices remain a strong buying opportunity. Remember that silver is significantly more volatile than gold and can retrace down closer to β‰ˆ$70, but it's definitely not the top for this bullrun in metals

Very excited for the upcoming price rebound in commodity metals

Current prices remain a strong buying opportunity.

Remember that silver is significantly more volatile than gold and can retrace down closer to β‰ˆ$70, but it's definitely not the top for this bullrun in metals
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

Random Forest Cat: An ML Trading Factory I open-sourced a multi-model, machine-learning based factory for trading strategies on GitHub. It's written in Python and is highly customizable, as it exposes a framework for configuring the pipeline via a DSL. Currently, it comes with

Random Forest Cat: An ML Trading Factory

I open-sourced a multi-model, machine-learning based factory for trading strategies on GitHub. It's written in Python and is highly customizable, as it exposes a framework for configuring the pipeline via a DSL.

Currently, it comes with
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

The problem with granular historical limit order book (LOB) data is its massive size I have a script downloading, decompressing and combining all of the data, and the total size was quickly over 300GB in size! I was hoping I could prototype & train a primary model from data in

The problem with granular historical limit order book (LOB) data is its massive size

I have a script downloading, decompressing and combining all of the data, and the total size was quickly over 300GB in size!

I was hoping I could prototype & train a primary model  from data in
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

Fingerprinting and loading β‰ˆ500GB of Bitcoin L1 LOB historical data from CSVs into a single DuckDB file database This will take a while, but then I will be able to train neural networks on that data πŸ˜„

Fingerprinting and loading β‰ˆ500GB of Bitcoin L1 LOB historical data from CSVs into a single DuckDB file database

This will take a while, but then I will be able to train neural networks on that data πŸ˜„
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

Using DuckDB To Train A Neural Network On 500GB Of Price Data I have β‰ˆ500GB of historical Bitcoin level 1 limit order book data to process and train a neural network on. I don't want to overcomplicate the data access layer and definitely want to keep all of the training,

Using DuckDB To Train A Neural Network On 500GB Of Price Data

I have β‰ˆ500GB of historical Bitcoin level 1 limit order book data to process and train a neural network on.

I don't want to overcomplicate the data access layer and definitely want to keep all of the training,
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

2 billion+ rows (and counting!) added to the DuckDB table containing Bitcoin's L1 LOB data so far The .duckdb file is now 32GB. This is going to be very interesting to process. If this experiment goes well -- I'll switch to DuckDB as the default storage layer for the pipeline

2 billion+ rows (and counting!) added to the DuckDB table containing Bitcoin's L1 LOB data so far

The .duckdb file is now 32GB. This is going to be very interesting to process. If this experiment goes well -- I'll switch to DuckDB as the default storage layer for the pipeline
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

Who would've thought that reordering 5 billion rows in a DuckDB table would need over 300GB in temporary storage on your hard drive Luckily there is `SET temp_directory`, so I'm running ORDER BY again pointing it to a medium that should meet DuckDB's storage needs I still

Who would've thought that reordering 5 billion rows in a DuckDB table would need over 300GB in temporary storage on your hard drive

Luckily there is `SET temp_directory`, so I'm running ORDER BY again pointing it to a medium that should meet DuckDB's storage needs

I still
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

Brent Crude Oil breaks above $90, and its headed higher! You can add this to my list of correct predictions. Last time oil was this expensive was more than 2 years ago in 2023. Perhaps it's obvious at this point -- but it won't top out at $90 πŸ˜„

Brent Crude Oil breaks above $90, and its headed higher!

You can add this to my list of correct predictions. Last time oil was this expensive was more than 2 years ago in 2023.

Perhaps it's obvious at this point -- but it won't top out at $90 πŸ˜„
Illya Gerasymchuk (httpz) (@illyagera) 's Twitter Profile Photo

UPDATE: the 5 billion row table reordering in DuckDB has been in progress for almost 24 hours The temporary directory used by the DuckDB process has filled up to almost 550GB πŸ˜… This would've been faster, but the temp storage has limited bandwidth

UPDATE: the 5 billion row table reordering in DuckDB has been in progress for almost 24 hours

The temporary directory used by the DuckDB process has filled up to almost 550GB πŸ˜…

This would've been faster, but the temp storage has limited bandwidth