Skip to content

feat: add production pattern notebooks (CrewAI 02, OpenAI Agents 02, LangChain 03)#63

Merged
shanjairaj7 merged 1 commit intomainfrom
feat/week2-production-notebooks
Mar 1, 2026
Merged

feat: add production pattern notebooks (CrewAI 02, OpenAI Agents 02, LangChain 03)#63
shanjairaj7 merged 1 commit intomainfrom
feat/week2-production-notebooks

Conversation

@shanjairaj7
Copy link
Copy Markdown
Collaborator

Summary

  • crewai_02_production_patterns.ipynb — thread persistence via SQLite ThreadRegistry, multi-inbox routing, deterministic idempotency keys, exponential backoff retry with dead-letter queue, and a webhook-ready ProductionEmailTool (18 cells)
  • openai_agents_02_patterns.ipynb — thread continuity via RunContextWrapper dataclass, extraction schemas with cost/latency comparison table, parallel asyncio agents, prompt injection defense with sanitize_for_llm + recipient domain validation (19 cells)
  • langchain_03_production.ipynb — singleton client pattern, search.threads() context building, delivery.metrics() health checks with alerting thresholds, delivery.suppressions() audit, FastAPI async webhook handler with signature verification, and sms.send() escalation for critical urgency emails (18 cells)

All notebooks have simulated output cells on every API-calling code cell (3-way semantic pair: markdown prose + code + output).

Test plan

  • Open each notebook in Colab to verify cell rendering
  • Confirm all code cells have simulated outputs
  • Verify no phantom API calls (search/sms/delivery all namespaced correctly)

…gChain

Three new notebooks covering week-2 production patterns:
- crewai_02_production_patterns: thread persistence via SQLite ThreadRegistry,
  multi-inbox routing, deterministic idempotency keys, exponential backoff
  retry with dead-letter queue, and a webhook-ready ProductionEmailTool
- openai_agents_02_patterns: thread continuity via RunContextWrapper dataclass,
  extraction schemas (cost/latency comparison), parallel asyncio agents,
  and prompt injection defense with sanitize_for_llm + recipient validation
- langchain_03_production: singleton client pattern, search.threads() context
  building, delivery.metrics() health checks with threshold alerting,
  delivery.suppressions() audit, FastAPI async webhook handler, and
  sms.send() escalation for critical urgency emails

All notebooks have simulated output cells on every API-calling code cell.
@shanjairaj7 shanjairaj7 merged commit 886fab8 into main Mar 1, 2026
0 of 3 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant