You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In this ticket, #596, we avoided that arbitrage trades involving just 1 market keep getting created. The question is, was the fix enough or do we need to make additional changes in the logic?
Check the trades from 01.01.2025 - 14.01.2025 and check for inconsistencies in the trades, namely
-> Trades involving the same market (main_market == related_market)
-> Any other reasons as to why the profitability of this agent is very low
The text was updated successfully, but these errors were encountered:
Hi @gabrielfior, apologies for jumping into the conversation without much context—I came across this discussion via the repository mentioned on Kernel Slack.
Do you currently use a monitoring setup, such as Prometheus with Grafana or native GCP tools (I noticed you’re using GCP)? I imagine it’s possible to track events like this using such systems. Ideally, we could store the data locally within the agent and expose it as a Prometheus metric, making these occurrences easier to identify and measure.
Thanks for your message @empeje .
We use Dune for tracking the trader agents activities -> https://dune.com/gnosischain_team/omen-ai-agents
We also use Langfuse for tracking the LLM calls to the cloud provider (for internal debugging reasons).
In this ticket, #596, we avoided that arbitrage trades involving just 1 market keep getting created. The question is, was the fix enough or do we need to make additional changes in the logic?
Check the trades from 01.01.2025 - 14.01.2025 and check for inconsistencies in the trades, namely
-> Trades involving the same market (main_market == related_market)
-> Any other reasons as to why the profitability of this agent is very low
The text was updated successfully, but these errors were encountered: