BEGIN:VCALENDAR VERSION:2.0 PRODID:-//HasGeek//NONSGML Funnel//EN DESCRIPTION:A hands-on workshop on designing short-term and long-term memo ry for reliable\, personalized AI agents X-WR-CALDESC:A hands-on workshop on designing short-term and long-term mem ory for reliable\, personalized AI agents NAME:From amnesia to photographic memory: building memory into agentic sys tems X-WR-CALNAME:From amnesia to photographic memory: building memory into age ntic systems REFRESH-INTERVAL;VALUE=DURATION:PT12H SUMMARY:From amnesia to photographic memory: building memory into agentic systems TIMEZONE-ID:Asia/Kolkata X-PUBLISHED-TTL:PT12H X-WR-TIMEZONE:Asia/Kolkata BEGIN:VEVENT SUMMARY:From amnesia to photographic memory: building memory into agentic systems DTSTART:20260808T053000Z DTEND:20260808T093000Z DTSTAMP:20260724T223138Z UID:session/RtVwMSzzUMNbvZ95iXhiYS@hasgeek.com SEQUENCE:5 CREATED:20260722T052441Z DESCRIPTION:## Objective\nYou have built a few agents\, wired them into a multi-agent system. What comes next? This workshop takes a focused look at one of the most critical and often underestimated parts of an agentic sys tem ā€” memory. Participants will come away with a practical understanding of how memory shapes agent behavior\, capabilities\, and user experience. \n\n**Format:** Hands-on workshop\n**Duration:** 4 hours\n\n---\n\n## Agen da\n### Introduction\nA quick grounding in agentic system fundamentals ā€” core components (LLM\, Tools\, Planning\, Execution\, Memory)\, the agent execution lifecycle (Observe\, Reason\, Act\, Reflect)\, and why memory i s a first-class concern rather than an afterthought.\n\n### Problems with a Stateless Agentic System\nWhat actually breaks when an agent has no stat e: loss of context within a conversation\, no personalisation\, repetitive interactions.\n\n**Hands-on Exercise 1:** Build a stateless travel assist ant\, run it through scenarios\, and identify what's missing.\n\n### Short -Term Memory\nTwo practical patterns: using workflow state to store prefer ences\, tool outputs\, and intermediate decisions\; and passing conversati on history in the prompt to maintain continuity.\n\nIncludes a section on managing conversation history efficiently ā€” what to preserve\, what to d rop\, and common strategies for keeping history useful without letting it balloon.\n\n### Memory Checkpoints\nCheckpoints\, why they're needed\, and the LangGraph-specific concepts required: checkpointers\, state persisten ce\, workflow resumption\, and interrupt-and-resume patterns.\n\n**Hands-o n Exercise 2:** Extend the travel assistant to retain budget\, interests\, destinations\, and trip duration across a session\, asking intelligent fo llow-ups and maintaining coherent context.\n\n### Long-Term Memory\nLimita tions of short-term memory\, when long-term memory becomes necessary\, and the three types ā€” semantic\, episodic\, and procedural. Architectures f or how memory is created\, stored\, and retrieved\, plus production best p ractices.\n\n**Hands-on Exercise 3:** Extend the assistant further so it r ecalls preferred destinations\, remembers budget ranges\, recommends based on past choices\, and picks up conversations across sessions.\n\n### Wrap -Up and Discussion\nReal trade-offs between memory approaches\, common pro duction challenges\, and Q&A.\n\n---\n\n## Key Takeaways\nBy the end of th is workshop\, participants will:\n\n- Understand the role of memory in age ntic systems\n- Build stateless\, session-aware\, and persistent agents\n- Implement short-term and long-term memory patterns\n- Use checkpoints eff ectively in workflow orchestration\n\n---\n\n## Who is this workshop for\n - Developers and engineers building AI-powered applications who want relia ble\, agent-driven workflows\n- ML/AI engineers exploring agentic architec tures\n- Tech leads and solution architects designing scalable\, multi-age nt systems and automation platforms\n- Platform and backend engineers inte grating LLMs with real-world systems\, APIs\, and infrastructure\n- Innova tion\, R&D\, and product teams experimenting with AI agents\, orchestratio n frameworks\, and MCP-based extensibility\n\nThe instructors will start w ith [this talk](https://youtu.be/zWWWi5tfkn4?si=SXbpzCFN62GDcAxa) to get t he fundamentals\, then join the workshop for a hands-on deep dive into mem ory architectures for agentic systems.\n\n---\n\n## About the workshop ins tructors\n**[Swetha A](https://www.linkedin.com/in/swetha0302)** is a soft ware developer and GenAI enthusiast\, Solution Consultant at Sahaj Softwar e. She builds intelligent systems using deep learning and generative AI\, with a research paper presented at the International Conference on Data An alytics and Management. Has delivered talks on AI agents and facilitated w orkshops on AI-assisted SDLC and multi-agent systems.\n**[Mahita D](https: //www.linkedin.com/in/mahita07)** is a software developer focused on agent ic systems and the evolving role of AI in software development. Shares exp eriences building agentic systems and helping teams turn AI capabilities i nto practical solutions through talks and workshops.\n\n---\n\n## How to a ttend this workshop\nThis workshop is open to:\nšŸŽŸļø Fifth Elephant com munity members ā€” https://hasgeek.com/fifthelephant#memberships\nšŸŽŸļø Ticket holders for The Fifth Elephant annual conference ā€” https://hasgee k.com/fifthelephant/enterprise-ai-in-production-meetup#tickets\n\n**This w orkshop is open to 30 participants (in-person) & hybrid access for remote attendees. Seats for in-person participants will be available on first-com e-first-served basis. šŸŽŸļø**\n\n---\n## Need more info?\nā˜Žļø Call: ( 91) 7676332020\nšŸ“§ Email: info@hasgeek.com LAST-MODIFIED:20260722T052542Z LOCATION:Sahaj Software\, Koramangala\, Bangalore - https://hasgeek.com/fi fthelephant/from-amnesia-to-photographic-memory-workshop/ ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com URL:https://hasgeek.com/fifthelephant/from-amnesia-to-photographic-memory- workshop/ BEGIN:VALARM ACTION:display DESCRIPTION:From amnesia to photographic memory: building memory into agen tic systems in 5 minutes TRIGGER:-PT5M END:VALARM END:VEVENT END:VCALENDAR