Skip to content

Improve AI runtime validation/retries, reporting, and data ingestion robustness #42

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Rohan5commit merged 2 commits into from
May 19, 2026
Merged

Improve AI runtime validation/retries, reporting, and data ingestion robustness #42

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
8c67d2a
Add AI config validation and runtime decision visibility
Rohan5commit May 19, 2026
f731ca8
Merge branch 'main' into codex/document-ai-bot-functionality-and-chan…
Rohan5commit May 19, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
76 changes: 69 additions & 7 deletions .github/workflows/daily_trading_bot.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -232,9 +232,47 @@ jobs:
python -c 'import json, os, pathlib; payload={"runtime_mode": os.environ["runtime_mode"], "selected_backend": os.environ["selected_backend"], "selected_compute_name": "", "selected_disk_gb": 0, "reason": os.environ["reason"], "preflight": {"router_policy": "cerebrium_primary_then_distilled"}}; pathlib.Path("results/ai_runtime_plan.json").write_text(json.dumps(payload, indent=2)+"\n", encoding="utf-8"); print(json.dumps(payload, indent=2))'
cat results/ai_runtime_plan.json


- name: Validate Cerebrium primary configuration
id: validate_cerebrium_config
if: ${{ steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) }}
run: |
configured_inference_url="${CEREBRIUM_INFERENCE_URL}"
if [ -z "${configured_inference_url}" ]; then
configured_inference_url="${CEREBRIUM_TRAINED_MODEL_URL}"
fi
if [ -z "${configured_inference_url}" ]; then
configured_inference_url="${TRAINED_MODEL_INFERENCE_URL}"
fi
if [ -n "${configured_inference_url}" ]; then
echo "has_inference_url=true" >> "$GITHUB_OUTPUT"
echo "inference_url_source=configured"
else
echo "has_inference_url=false" >> "$GITHUB_OUTPUT"
echo "inference_url_source=missing"
echo "AI primary inference URL is missing across CEREBRIUM_INFERENCE_URL/CEREBRIUM_TRAINED_MODEL_URL/TRAINED_MODEL_INFERENCE_URL" >&2
fi
{
echo "runtime_mode=${runtime_mode}"
echo "selected_backend=${selected_backend}"
echo "selected_compute_name="
echo "selected_disk_gb=0"
echo "reason=${reason}"
} >> "$GITHUB_OUTPUT"
python -c 'import json, os, pathlib; payload={"runtime_mode": os.environ["runtime_mode"], "selected_backend": os.environ["selected_backend"], "selected_compute_name": "", "selected_disk_gb": 0, "reason": os.environ["reason"], "preflight": {"router_policy": "cerebrium_primary_then_distilled"}}; pathlib.Path("results/ai_runtime_plan.json").write_text(json.dumps(payload, indent=2)+"\n", encoding="utf-8"); print(json.dumps(payload, indent=2))'
cat results/ai_runtime_plan.json

- name: Emit AI runtime decision
if: ${{ steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) }}
run: |
echo "AI runtime mode: ${{ steps.plan_ai_runtime.outputs.runtime_mode }}"
echo "AI backend: ${{ steps.plan_ai_runtime.outputs.selected_backend }}"
echo "AI reason: ${{ steps.plan_ai_runtime.outputs.reason }}"
echo "AI inference URL configured: ${{ steps.validate_cerebrium_config.outputs.has_inference_url }}"

- name: Warm Cerebrium inference app
id: warm_cerebrium_inference
if: ${{ steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' }}
if: ${{ steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' && steps.validate_cerebrium_config.outputs.has_inference_url == 'true' }}
continue-on-error: true
timeout-minutes: 45
env:
Expand All @@ -255,7 +293,7 @@ jobs:

- name: Verify Cerebrium predict endpoint
id: verify_cerebrium_predict
if: ${{ steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' }}
if: ${{ steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' && steps.validate_cerebrium_config.outputs.has_inference_url == 'true' }}
continue-on-error: true
timeout-minutes: 3
env:
Expand Down Expand Up @@ -316,7 +354,7 @@ jobs:

- name: Run AI Trading Bot on Cerebrium
id: run_ai_bot_cerebrium
if: ${{ steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' && steps.verify_cerebrium_predict.outcome == 'success' }}
if: ${{ steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' && steps.validate_cerebrium_config.outputs.has_inference_url == 'true' && steps.verify_cerebrium_predict.outcome == 'success' }}
continue-on-error: true
timeout-minutes: 90
env:
Expand Down Expand Up @@ -347,7 +385,7 @@ jobs:

- name: Run AI Trading Bot on Cerebrium (Retry)
id: run_ai_bot_cerebrium_retry
if: ${{ always() && steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' && steps.verify_cerebrium_predict.outcome == 'success' && steps.run_ai_bot_cerebrium.outcome == 'failure' }}
if: ${{ always() && steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' && steps.validate_cerebrium_config.outputs.has_inference_url == 'true' && steps.verify_cerebrium_predict.outcome == 'success' && steps.run_ai_bot_cerebrium.outcome == 'failure' }}
continue-on-error: true
timeout-minutes: 90
env:
Expand Down Expand Up @@ -444,7 +482,7 @@ jobs:

- name: Run AI Trading Bot (Distilled Local Fallback)
id: run_ai_bot_distilled_local
if: ${{ steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && (steps.plan_ai_runtime.outputs.runtime_mode == 'distilled_local' || (steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' && (steps.warm_cerebrium_inference.outcome == 'failure' || steps.verify_cerebrium_predict.outcome == 'failure' || (steps.run_ai_bot_cerebrium.outcome == 'failure' && steps.run_ai_bot_cerebrium_retry.outcome != 'success'))) || (steps.plan_ai_runtime.outputs.runtime_mode == 'lightning_full' && (steps.launch_lightning_inference.outcome == 'failure' || steps.run_ai_bot_in_lightning_studio.outcome == 'failure'))) }}
if: ${{ steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && (steps.plan_ai_runtime.outputs.runtime_mode == 'distilled_local' || (steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' && steps.validate_cerebrium_config.outputs.has_inference_url == 'true' && (steps.warm_cerebrium_inference.outcome == 'failure' || steps.verify_cerebrium_predict.outcome == 'failure' || (steps.run_ai_bot_cerebrium.outcome == 'failure' && steps.run_ai_bot_cerebrium_retry.outcome != 'success'))) || (steps.plan_ai_runtime.outputs.runtime_mode == 'lightning_full' && (steps.launch_lightning_inference.outcome == 'failure' || steps.run_ai_bot_in_lightning_studio.outcome == 'failure'))) }}
continue-on-error: true
timeout-minutes: 60
env:
Expand All @@ -459,7 +497,31 @@ jobs:
LOG_LEVEL: ${{ secrets.LOG_LEVEL }}
DISABLE_CORE_TRADING: "1"
AI_RUNTIME_MODE: "distilled_local"
AI_ROUTER_REASON: ${{ steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' && 'distilled_after_cerebrium_failure' || (steps.plan_ai_runtime.outputs.runtime_mode == 'lightning_full' && 'distilled_after_lightning_failure' || steps.plan_ai_runtime.outputs.reason) }}
AI_ROUTER_REASON: ${{ steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' && steps.validate_cerebrium_config.outputs.has_inference_url == 'true' && 'distilled_after_cerebrium_failure' || (steps.plan_ai_runtime.outputs.runtime_mode == 'lightning_full' && 'distilled_after_lightning_failure' || steps.plan_ai_runtime.outputs.reason) }}
run: |
python quant_platform/scripts/prepare_ai_only_retry.py --results-dir results
python main.py daily_job

- name: Run AI Trading Bot (Emergency Distilled Retry)
id: run_ai_bot_distilled_retry
if: ${{ always() && steps.install_base_dependencies.outcome == 'success' && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && steps.run_ai_bot_distilled_local.outcome == 'failure' }}
continue-on-error: true
timeout-minutes: 60
env:
SMTP_SERVER: ${{ secrets.SMTP_SERVER }}
SMTP_PORT: ${{ secrets.SMTP_PORT }}
SENDER_EMAIL: ${{ secrets.SENDER_EMAIL }}
SENDER_PASSWORD: ${{ secrets.SENDER_PASSWORD }}
RECIPIENT_EMAIL: ${{ secrets.RECIPIENT_EMAIL }}
TRAINED_MODEL_API_KEY: ${{ secrets.TRAINED_MODEL_API_KEY }}
TWELVEDATA_API_KEYS: ${{ secrets.TWELVEDATA_API_KEYS }}
ARM_LIVE_TRADING: ${{ secrets.ARM_LIVE_TRADING }}
LOG_LEVEL: ${{ secrets.LOG_LEVEL }}
DISABLE_CORE_TRADING: "1"
AI_RUNTIME_MODE: "distilled_local"
AI_ROUTER_REASON: "distilled_emergency_retry_after_failure"
TRAINED_MODEL_TIMEOUT_SECONDS: "450"
TRAINED_MODEL_MAX_RETRIES: "1"
run: |
python quant_platform/scripts/prepare_ai_only_retry.py --results-dir results
python main.py daily_job
Expand Down Expand Up @@ -489,7 +551,7 @@ jobs:
python main.py daily_job

- name: Send AI Failure Report
if: ${{ always() && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && (((steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full') && (steps.warm_cerebrium_inference.outcome == 'failure' || steps.verify_cerebrium_predict.outcome == 'failure' || (steps.run_ai_bot_cerebrium.outcome == 'failure' && steps.run_ai_bot_cerebrium_retry.outcome != 'success')) && steps.run_ai_bot_distilled_local.outcome == 'failure' && steps.run_ai_bot_distilled_retry.outcome != 'success') || ((steps.plan_ai_runtime.outputs.runtime_mode == 'lightning_full') && (steps.launch_lightning_inference.outcome == 'failure' || steps.run_ai_bot_in_lightning_studio.outcome == 'failure') && steps.run_ai_bot_distilled_local.outcome == 'failure' && steps.run_ai_bot_distilled_retry.outcome != 'success') || ((steps.plan_ai_runtime.outputs.runtime_mode == 'distilled_local') && steps.run_ai_bot_distilled_local.outcome == 'failure' && steps.run_ai_bot_distilled_retry.outcome != 'success')) }}
if: ${{ always() && !(github.event_name == 'workflow_dispatch' && inputs.run_in_lightning_studio) && ((github.event_name != 'workflow_dispatch') || (github.event_name == 'workflow_dispatch' && !inputs.disable_ai_trading)) && (((steps.plan_ai_runtime.outputs.runtime_mode == 'cerebrium_full' && steps.validate_cerebrium_config.outputs.has_inference_url == 'true') && (steps.warm_cerebrium_inference.outcome == 'failure' || steps.verify_cerebrium_predict.outcome == 'failure' || (steps.run_ai_bot_cerebrium.outcome == 'failure' && steps.run_ai_bot_cerebrium_retry.outcome != 'success')) && steps.run_ai_bot_distilled_local.outcome == 'failure' && steps.run_ai_bot_distilled_retry.outcome != 'success') || ((steps.plan_ai_runtime.outputs.runtime_mode == 'lightning_full') && (steps.launch_lightning_inference.outcome == 'failure' || steps.run_ai_bot_in_lightning_studio.outcome == 'failure') && steps.run_ai_bot_distilled_local.outcome == 'failure' && steps.run_ai_bot_distilled_retry.outcome != 'success') || ((steps.plan_ai_runtime.outputs.runtime_mode == 'distilled_local') && steps.run_ai_bot_distilled_local.outcome == 'failure' && steps.run_ai_bot_distilled_retry.outcome != 'success')) }}
continue-on-error: true
env:
SMTP_SERVER: ${{ secrets.SMTP_SERVER }}
Expand Down