Resolution Revolution: Turbocharging Security Ticketing Timelines
summary
An Adobe security AI team builds a unified knowledge base that merges bug details, code snippets and expert notes to cut ticket turnaround times; it processes ~15–20k historical tickets into vectors and surfaces representative code fixes rather than automatic PRs, an expert finder that links past ticket authors, and best-practice RAG outputs. Stack: Azure OpenAI, LangChain/LangGraph and Pinecone vector DB. Results: prototype delivers actionable guidance, reduces lookup time, keeps humans in loop, and flags when hallucination or missing context prevents an automatic fix.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 1:09Security challenges surge
- 5:49AI code fixing pitfalls
- 10:29Unified knowledge base
- 12:17Coder configuration fixes
- 14:48Finding experts faster
- 17:21Scaling security guidance
- 21:37Leveraging similar tickets
- 25:10Evaluating AI outputs
- 27:59Key learnings
- 31:29Takeaways and AI empowerment
- 33:58Q&A
related talks
09mThe AI trick that cut AWS's issue triage time by 90% and caught regressions in 2 minutes
29mOne-Click Code Fix: Securing Code Using AI
30mShip Reliable RAG: Evals That Grow with Your Retrieval Complexity
46mAgentic Engineering: From Writing Code to Reviewing It
50mSpeeding Up Vulnerability Triage: Automating Context Retrieval with AI Agents
44mPanel Discussion | The AI Revolution in Testing: Opportunities and Challenges
01h 39mShipping complex AI applications
23mCode Security Reinvented Navigating the era of AI