BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//HasGeek//NONSGML Funnel//EN
DESCRIPTION:Open Source AI meet-up - Hyderabad edition
X-WR-CALDESC:Open Source AI meet-up - Hyderabad edition
NAME:Software Life Cycle Development and AI
X-WR-CALNAME:Software Life Cycle Development and AI
REFRESH-INTERVAL;VALUE=DURATION:PT12H
SUMMARY:Software Life Cycle Development and AI
TIMEZONE-ID:Asia/Kolkata
X-PUBLISHED-TTL:PT12H
X-WR-TIMEZONE:Asia/Kolkata
BEGIN:VEVENT
SUMMARY:Check-in - on first-come-first-serve basis
DTSTART:20251101T042500Z
DTEND:20251101T044500Z
DTSTAMP:20260419T202815Z
UID:session/EckCu7Np6uVFhaLdqjsJF@hasgeek.com
SEQUENCE:4
CREATED:20251009T042053Z
LAST-MODIFIED:20251031T173500Z
LOCATION:Thoughtworks\, Hyderabad
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
BEGIN:VALARM
ACTION:display
DESCRIPTION:Check-in - on first-come-first-serve basis in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Introduction to the meet-up - theme and schedule
DTSTART:20251101T044500Z
DTEND:20251101T050000Z
DTSTAMP:20260419T202815Z
UID:session/XUyPwyveA3kXCtTPwD1GFr@hasgeek.com
SEQUENCE:5
CREATED:20251009T042129Z
GEO:17.422;78.339
LAST-MODIFIED:20251026T151744Z
LOCATION:Fusion Space - Thoughtworks Technologies (India) Pvt Ltd\nHyderab
 ad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
BEGIN:VALARM
ACTION:display
DESCRIPTION:Introduction to the meet-up - theme and schedule in Fusion Spa
 ce in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Sponsored keynote: AI-first software delivery: superpowers\, adopt
 ion challenges\, and the path to software 3.0
DTSTART:20251101T050000Z
DTEND:20251101T054500Z
DTSTAMP:20260419T202815Z
UID:session/5NgsrKyCWSJszHXvHKHkdW@hasgeek.com
SEQUENCE:3
CATEGORIES:Sponsored talk (25-35 mins)
CREATED:20251021T125318Z
DESCRIPTION:AI is no longer a sidekick in the software delivery process\, 
 it’s becoming an integral partner that reshapes how we ideate\, build\, 
 and operate digital products. This talk explores how AI’s infusion acros
 s the software value chain—from AI-assisted requirements analysis and de
 sign to AI-augmented coding\, testing\, and AIOps—is ushering in a new e
 ra of software 3.0. Drawing from Thoughtworks’ global experiences\, we
 ’ll examine how organizations can move beyond “copilot thinking” to 
 architect for AI-First Software Delivery.\n\nStory points and cycle time a
 lone are anemic in measuring the impact of infusing AI into the software v
 alue chain\, while vanity metrics such as “lines of code produced by AI
 ” can be misleading. We’ll reframe how to measure what truly matters
 —across productivity and flow\, quality and resilience\, and value and i
 nnovation—as teams adapt to AI-driven delivery. We shall also explore wh
 ere we are heading in terms of software 3.0.\n\n**Key takeaways:**\n\n1. U
 nderstand the shift from AI-assisted to AI-first software delivery.\n2. Di
 scover the superpowers of Generative AI and how they transform every stage
  of the software value chain.\n3. Explore meaningful metrics that reflect 
 value\, quality\, and team flow in AI-augmented delivery.\n4. Apply pragma
 tic strategies for embedding AI safely and effectively in real-world envir
 onments.\n\n**Bio**\nVanya is an experienced technologist with a demonstra
 ted history of working in the IT services industry. She is a strategic con
 sultant specializing in data mesh\, platforms\, delivery infrastructure\, 
 cloud native evolutionary architecture\, and AI first software delivery.\n
 \nIn her current role as Head of Technology for Thoughtworks in India\, Va
 nya shapes new pursuits\, strategically advises on client projects\, and f
 ormulates meaningful and resilient technology strategies. She is also part
  of the Thoughtworks' Technology Advisory Board that puts together thought
 s on emerging technology trends in the industry\, resulting in the much so
 ught after [Technology Radar](https://www.thoughtworks.com/radar?utm_sourc
 e=website&utm_medium=twerprofile&utm_campaign=tech-radar-india-launch).\n\
 nVanya is a passionate technologist with a knack for solving complex probl
 ems at scale. She is also deeply involved with open-source communities.\n
GEO:17.422;78.339
LAST-MODIFIED:20251021T125407Z
LOCATION:Fusion Space - Thoughtworks Technologies (India) Pvt Ltd\nHyderab
 ad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
URL:https://hasgeek.com/fifthelephant/osai-hyd-meetup/schedule/ai-first-so
 ftware-delivery-superpowers-adoption-challenges-and-the-path-to-software-3
 -0-5NgsrKyCWSJszHXvHKHkdW
BEGIN:VALARM
ACTION:display
DESCRIPTION:Sponsored keynote: AI-first software delivery: superpowers\, a
 doption challenges\, and the path to software 3.0 in Fusion Space in 5 min
 utes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Self-updating knowledge bases for smarter AI agents
DTSTART:20251101T054500Z
DTEND:20251101T063000Z
DTSTAMP:20260419T202815Z
UID:session/TCiFxESfizz3c1ss5J9ZxS@hasgeek.com
SEQUENCE:6
CATEGORIES:25-35 min talk
CREATED:20251009T042239Z
DESCRIPTION:## Abstract\nAs we begin to rely more on AI agents to assist u
 s\, it is also important to realize that they are only as useful as the kn
 owledge they have access to.\n\nA substantial amount of modern development
  relies on external integrations\, and developers make heavy use of docume
 ntation\, ranging from precise technical API specifications to verbose\, t
 ext-heavy tutorials. With a wide array of integrations and an unpredictabl
 e pace of changes\, developers have to constantly track changes and incorp
 orate them into their work.\n\nTraditional document management systems wil
 l often fall short in keeping pace with the evolving APIs\, their specs\, 
 and developer guides. It goes without saying that we should be leveraging 
 AI agents to make the lives of developers easier here. But\, agents are no
 t magical entities. They perform only as well as the instructions and cont
 ext they are fed. And that is **our** responsibility.\n\nIn this talk\, we
  present an approach that combines:\n\n* Intelligent web crawling  \n* Sem
 antic search  \n* Model Context Protocol (MCP)  \n* Change Impact Analysis
  and Context Structuring: *extract structured\, actionable context for AI 
 agents from documentation diffs*\n\nto create a precise\, well-structured 
 system that agents can use to help developers keep their code integrations
  up-to-date with changes in third-party systems. This system provides the 
 ability to detect and act on API contract changes\, such as deprecations a
 nd schema modifications that introduce new\, salient features. Instead of 
 merely flagging these changes\, the system can automatically contextualize
  them and highlight the exact places that need updates. This granular\, ve
 rsioned information directly feeds into AI workflows\, enabling timely\, p
 roactive adjustments to code and logic.\n\nThe session will include demons
 trations of real-time documentation monitoring and AI agent adaptation to 
 evolving APIs\, offering concrete examples of the approach in practice. At
 tendees will leave with actionable insights into designing resilient\, ada
 ptive AI systems capable of handling real-world complexity.\n\n## Key Take
 aways\n\n* The challenges of documentation drift and its impact on AI syst
 em reliability  \n* Designing and implementing self-updating knowledge bas
 es using intelligent crawling\, semantic search\, and versioned change tra
 cking  \n* Leveraging Model Context Protocol (MCP) for precise\, contextua
 lized change detection in APIs and documentation  \n* Practical demonstrat
 ions of AI systems responding to real-time changes\, with clear visibility
  into what has changed and how  \n* Best practices for building robust\, s
 elf-adapting AI workflows in production environments\n## Target Audience\n
 \n* Engineering teams using AI-assisted development tools  \n* Platform En
 gineers managing multiple API integrations  \n* Development leaders seekin
 g to eliminate documentation drift as a source of technical debt.\n\n**Fif
 th-EL BLR Lightning Talk**: [https://www.youtube.com/live/4Nw89rICm-s?feat
 ure=shared&t=18585](https://www.youtube.com/live/4Nw89rICm-s?feature=share
 d&t=18585)\n**Open Source Link**: [https://github.com/iamakash-06/Doc-Moni
 tor-MCP](https://github.com/iamakash-06/Doc-Monitor-MCP)  \n **Reference L
 ink**: [https://sahaj.ai/mcp-and-doc-monitor-transform-external-services-i
 ntegration/](https://sahaj.ai/mcp-and-doc-monitor-transform-external-servi
 ces-integration/)\n**Draft Slides**: [https://docs.google.com/presentation
 /d/1cpI83sa7rfvJBs50dS00GBQH1uvbUiUm9-Vs7s2fqfE/edit?usp=sharing](https://
 docs.google.com/presentation/d/1cpI83sa7rfvJBs50dS00GBQH1uvbUiUm9-Vs7s2fqf
 E/edit?usp=sharing)\n\n## Speaker Bio\nAkash Sathish is a Solution Consult
 ant at Sahaj Software and a machine learning enthusiast with hands-on expe
 rience in AI-assisted solution delivery. He has delivered talks in collabo
 ration with GitTogether\, DeepLearning.AI and Intel. Known for his knack f
 or storytelling\, Akash enjoys breaking down complex concepts into simple\
 , relatable insights and making advanced topics accessible to a wide range
  of audiences.\n\nProfile Links and Previous Talk links:\n\n1. [LinkedIn](
 https://www.linkedin.com/in/akash-s-74a4271ba/)\n2. [The Art of Pairing wi
 th AI - Sahaj DevDay](https://youtu.be/QiqkyKOtq7c?feature=shared)  \n3. [
 Privacy Preserving Machine Learning \\- Sahaj DevDay](https://youtu.be/TQf
 gl3TO7RE?si=087Fv8DpokMjBmxn)   \n4. [Breaking Into AIoT \\- Intel IoT Clu
 b in Collaboration with Deep Learning.AI](https://youtu.be/0KvgVYkIwiI?si=
 8eDw6sPy9n3MOyWc&t=4794)   \n5. [Breaking Into AI \\- Intel IoT Club](http
 s://youtu.be/9EqOxlXgTiU?si=LqHjaq6wXGyP-fjK&t=1189)
GEO:17.422;78.339
LAST-MODIFIED:20251026T152007Z
LOCATION:Fusion Space - Thoughtworks Technologies (India) Pvt Ltd\nHyderab
 ad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
URL:https://hasgeek.com/fifthelephant/osai-hyd-meetup/schedule/building-se
 lf-updating-knowledge-bases-for-ai-agents-with-intelligent-monitoring-usin
 g-mcp-TCiFxESfizz3c1ss5J9ZxS
BEGIN:VALARM
ACTION:display
DESCRIPTION:Self-updating knowledge bases for smarter AI agents in Fusion 
 Space in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Sponsored demo: Comprehensive code security - beyond traditional r
 ules
DTSTART:20251101T063000Z
DTEND:20251101T065500Z
DTSTAMP:20260419T202815Z
UID:session/5a7YpCTawAj37KjKGXtFN2@hasgeek.com
SEQUENCE:4
CATEGORIES:Sponsored pitch (10 mins)
CREATED:20251021T125427Z
DESCRIPTION:As modern applications evolve\, static\, rule-based security i
 s no longer enough. Flyingduck introduces an intelligent\, developer-first
  approach to code security that transcends traditional methods. By combini
 ng SAST\, SCA\, SBOM\, Secrets Detection\, and Business Logic Flaw analysi
 s under one unified platform\, Flyingduck delivers context-aware insights 
 that focus on real\, exploitable risks rather than noise. Powered by AI-dr
 iven reachability and deep logic analysis\, it helps teams detect\, priori
 tize\, and fix vulnerabilities faster — directly within their developmen
 t workflows. Seamless integrations with GitHub\, GitLab\, Jira\, and Slack
  make secure coding an effortless part of everyday development. Flyingduck
  enables organizations to move beyond rules — towards truly comprehensiv
 e\, continuous\, and intelligent code security.\n\n**Bio:** Satheesh Chall
 a is Co-Founder and Chief Technology Officer at Flyingduck\, where he over
 sees the development of cloud-native security solutions. He leads the desi
 gn of scalable platforms\, AI-driven vulnerability analysis\, and ensures 
 compliance with standards such as GDPR\, PCI\, HIPAA\, ISO\, and SOC. His 
 work focuses on building reliable\, secure\, and efficient systems for mod
 ern development teams.
GEO:17.422;78.339
LAST-MODIFIED:20251026T125908Z
LOCATION:Fusion Space - Thoughtworks Technologies (India) Pvt Ltd\nHyderab
 ad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
URL:https://hasgeek.com/fifthelephant/osai-hyd-meetup/schedule/comprehensi
 ve-code-security-beyond-traditional-rules-5a7YpCTawAj37KjKGXtFN2
BEGIN:VALARM
ACTION:display
DESCRIPTION:Sponsored demo: Comprehensive code security - beyond tradition
 al rules in Fusion Space in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Birds of Feather (BOF) session: Choosing wisely: when AI agents ma
 ke sense
DTSTART:20251101T063000Z
DTEND:20251101T073000Z
DTSTAMP:20260419T202815Z
UID:session/H8rR1c8asEV5Y9HToHFifF@hasgeek.com
SEQUENCE:19
CATEGORIES:Birds of Feather (BOF) session 
CREATED:20251009T042846Z
DESCRIPTION:# Choosing Wisely: When AI Agents Make Sense\n\nAgentic AI is 
 one of today’s fastest-growing trends—and also one of the most misunde
 rstood.  \nHow can we separate real opportunities for agents from the hype
 ?  \nHow do we decide when an AI agent adds value versus when a simple rul
 e-based or traditional ML system is enough?\n\nJoin this **Birds-of-a-Feat
 her** session to find out!\n\n---\n\n## Potential Topics\n\n- **What defin
 es an AI agent**  \n  Interpreting context\, reasoning\, planning\, taking
  actions\, making decisions\, and adapting to change.\n\n- **Evaluating th
 e need for agentic capabilities**  \n  How to determine if these capabilit
 ies are truly required for a problem.\n\n- **Signals for simpler solutions
 **  \n  Indicators that a task is better served by traditional programming
  or ML.\n\n- **Pitfalls of “force-fitting” agents**  \n  Risks of inse
 rting agents into deterministic workflows where they add little value.\n\n
 - **Real-world examples**  \n  Cases of both successful and unnecessary us
 es of AI agents.\n\n---\n\n## Key Takeaways\n\n- A **practical framework**
  to decide when an AI agent is the right solution.\n- **Strategies to bala
 nce innovation with reliability** when integrating agents.\n- Lessons lear
 ned from real-world cases where agents **excel—or fail**.\n\n---\n\n## W
 ho Should Attend\n\n- **Engineers\, product managers\, architects\, and de
 cision-makers** evaluating AI adoption.\n- **Technical leaders** looking t
 o distinguish genuine agentic value from hype.\n- **Practitioners** eager 
 to share or learn best practices for deploying agents responsibly.\n\n## A
 bout Me\nAbout Me:\n\nI work as a Solution Consultant at Sahaj.ai. \n\nAt 
 Sahaj\, I am learning to build technology platforms with extreme programmi
 ng practice and work in high-trust and accountability technology teams.\n\
 nI am also working as Office Lead for Sahaj Hyderabad Office with purpose 
 to build community of Sahajeevis in Hyderabad with Strong tech and consult
 ing culture the Sahaj way\n\nI have 16 years of Software Engineering exper
 ience focussed on Tech Consulting and implementation and building strong t
 echnology teams. I have worked on Distributed systems\, and traditional mo
 nolith platforms and built products on top of Eclipse Platform. I have div
 erse experience in Software Engineering - building tech platforms dealing 
 with PII Data and requiring regulatory compliance\, delivery management\, 
 customer relationship management\, and presales.\n\nLinkedIn Profile:\nhtt
 ps://www.linkedin.com/in/priyadarshanpatil/
GEO:17.422;78.339
LAST-MODIFIED:20251028T065143Z
LOCATION:Abdul Kalam training room - Thoughtworks Technologies (India) Pvt
  Ltd\nHyderabad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
URL:https://hasgeek.com/fifthelephant/osai-hyd-meetup/schedule/choosing-wi
 sely-when-ai-agents-make-sense-H8rR1c8asEV5Y9HToHFifF
BEGIN:VALARM
ACTION:display
DESCRIPTION:Birds of Feather (BOF) session: Choosing wisely: when AI agent
 s make sense in Abdul Kalam training room in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Spec Driven Development (SDD): AI builds from your specs
DTSTART:20251101T065500Z
DTEND:20251101T073000Z
DTSTAMP:20260419T202815Z
UID:session/CwmwdeTqZy4LQKs6w9v7Ek@hasgeek.com
SEQUENCE:12
CATEGORIES:25-35 min talk
CREATED:20251009T042759Z
DESCRIPTION:Talk Abstract\n\nWhile chat-based AI copilots have revolutioni
 zed coding in 2024-25\, they still require developers to think in implemen
 tation details. This talk introduces Spec Driven Development (SDD) - a par
 adigm shift where developers write comprehensive specifications that gener
 ate entire systems. Through live demonstration\, we'll build a real applic
 ation using SDD\, showing how this approach delivers 10x productivity gain
 s over current AI-assisted development methods.\n\nTalk Outline\nPart 1: T
 he Evolution Gap (5 minutes)\n- Quick journey: Waterfall → Agile → Dev
 Ops → AI Copilots\n- Current reality: Chat copilots (Copilot\, Cursor\, 
 Claude) still require line-by-line thinking\n- The problem: We're using AI
  as a faster typist\, not as a true development -partner\n- Key insight: T
 he bottleneck has shifted from typing speed to specification clarity\n\nPa
 rt 2: Introducing Spec Driven Development (5 minutes)\n- What is SDD? Writ
 e what you want\, not how to build it\n- The paradigm shift:\n-Traditional
 : Write code → Test → Deploy\n-Chat Copilots: Describe code → AI ass
 ists → Review → Deploy\n-SDD: Specify system → AI implements → Val
 idate → Deploy\n- Real difference: Moving from imperative instructions t
 o declarative specifications\n- Why now? Convergence of LLMs\, formal veri
 fication\, and automated testing\n\nPart 3: Core Tenets of SDD (7 minutes)
 \n- Specification as Source of Truth\n  -The spec IS the application\n  -V
 ersion control tracks spec changes\, not code changes\n- Continuous Valida
 tion\n  -Real-time verification between spec and implementation\n  -Automa
 tic test generation from specifications\n- Implementation Agnosticism\n-Sa
 me spec can generate Python\, Java\, or Go implementations\n-Framework dec
 isions become deployment-time choices\n- Human-Centric Design\n-Natural la
 nguage with formal precision\n-Domain experts can write specs without codi
 ng knowledge\n\nPart 4: Live Demo - Building a Real Application (10 minute
 s)\nDemo: Creating a Rate Limiter Service\n- Writing the Spec (2 mins)\n-N
 atural language specification\n-Constraints and requirements\n-Performance
  criteria\n- Generation Phase (2 mins)\n-Watch AI generate complete implem
 entation\n-Multiple language outputs from same spec\n-Automatic test suite
  creation\n- Validation & Refinement (3 mins)\n-Live validation against sp
 ec\n-Adjusting spec based on requirements\n-Regeneration with improvements
 \n- Deployment (3 mins)\n-Container generation\n-API documentation auto-cr
 eated\n-Performance benchmarks vs hand-written code\n\n  Part 5: Practical
  Adoption (3 minutes)\n- Start small: Use SDD for new microservices\n- Int
 egration patterns with existing codebases\n- Tools available today vs what
 's coming\n- Skills developers need to develop\n- Call to action: Join the
  SDD community and early adopter program\n\nKey Takeaways\n- SDD is not ab
 out replacing developers but elevating them to system designers\n- Practic
 al steps to implement SDD in your organization today\n- Understanding why 
 specification quality becomes the new competitive advantage\n- Hands-on ex
 perience seeing a complete system generated from specification\n\nWhy This
  Talk Matters for The Fifth Elephant\nThe Fifth Elephant audience consists
  of engineers building complex\, scalable systems. SDD is particularly pow
 erful for:\n- Microservices: Generate entire service meshes from architect
 ural specs\n- Data Pipelines: Specify data flow\, let AI handle implementa
 tion\n- ML Infrastructure: Focus on model requirements\, not boilerplate c
 ode\n- Scale: Build distributed systems that would take months in days\n\n
 Unique Value Proposition\nThis isn't theoretical - attendees will witness 
 a live\, working application being built from specification to deployment 
 in under 10 minutes. They'll leave with:\n- Access to SDD tools and framew
 orks\n- Template specifications for common patterns\n- Clear roadmap for o
 rganizational adoption\n- Community access for continued learning\n\nTarge
 t Audience\n- Senior developers and architects\n- Engineering managers and
  technical leads\n- CTOs and technology decision makers\n- Anyone frustrat
 ed with the pace of current development\n\nDemo Backup Plan\n- If live dem
 o fails\, I have:\n- Pre-recorded video of the same demo\n- Screenshots of
  each step\n- Generated code examples ready to display\n\nSpeaker Informat
 ion\nName: Hitesh Kumar\, Joe Noel\nOrganization: KnackLabs\nBio: Hitesh a
 nd Joe are senior AI developers with deep expertise in software developmen
 t methodologies\, programming paradigms\, and the evolution of human-AI co
 llaboration in software engineering. Passionate about advancing developmen
 t practices that enhance productivity while maintaining code quality and d
 eveloper satisfaction.
GEO:17.422;78.339
LAST-MODIFIED:20251028T065229Z
LOCATION:Fusion Space - Thoughtworks Technologies (India) Pvt Ltd\nHyderab
 ad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
URL:https://hasgeek.com/fifthelephant/osai-hyd-meetup/schedule/beyond-chat
 -copilots-the-future-of-development-with-spec-driven-development-sdd-Cwmwd
 eTqZy4LQKs6w9v7Ek
BEGIN:VALARM
ACTION:display
DESCRIPTION:Spec Driven Development (SDD): AI builds from your specs in Fu
 sion Space in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Lunch
DTSTART:20251101T073000Z
DTEND:20251101T083000Z
DTSTAMP:20260419T202815Z
UID:session/YL3VeS9QnG2umqvv7t4NBg@hasgeek.com
SEQUENCE:8
CREATED:20251009T042353Z
GEO:17.422;78.339
LAST-MODIFIED:20251028T065153Z
LOCATION:Abdul Kalam training room - Thoughtworks Technologies (India) Pvt
  Ltd\nHyderabad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
BEGIN:VALARM
ACTION:display
DESCRIPTION:Lunch in Abdul Kalam training room in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Lunch
DTSTART:20251101T073000Z
DTEND:20251101T083000Z
DTSTAMP:20260419T202815Z
UID:session/DbspiB28codb1DKLuNTdG9@hasgeek.com
SEQUENCE:9
CREATED:20251009T042902Z
GEO:17.422;78.339
LAST-MODIFIED:20251028T065239Z
LOCATION:Fusion Space - Thoughtworks Technologies (India) Pvt Ltd\nHyderab
 ad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
BEGIN:VALARM
ACTION:display
DESCRIPTION:Lunch in Fusion Space in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:DSPy: building scalable\, production-ready LLM applications
DTSTART:20251101T083000Z
DTEND:20251101T091500Z
DTSTAMP:20260419T202815Z
UID:session/6QtnmbdmErb6y2PqssmtRv@hasgeek.com
SEQUENCE:12
CATEGORIES:25-35 min talk
CREATED:20251009T045023Z
DESCRIPTION:# Description\nThe rapid evolution of Large Language Models (L
 LMs) has introduced a paradigm shift in AI development\, but building reli
 able\, production ready applications remains a challenge. Developers often
  face brittle systems\, driven by ad hoc prompting and rapidly changing mo
 del guides. This session introduces DSPy\, an open source\, lightweight\, 
 declarative framework that brings engineering discipline back into the pro
 cess\, transforming LLM development from chaotic prompt engineering into s
 tructured\, maintainable programming.\n\nIn this talk\, we will start with
  the basics of DSPy’s core abstractions\, Signatures and Modules\, and d
 emonstrate how they help define tasks declaratively and encapsulate infere
 nce strategies. We’ll then progressively enhance these applications for 
 production by introducing adapters\, multimodality support\, robust module
 s with scalability features\, and DSPy Optimizers like MIPROv2 and SIMBA\,
  that improve performance. We will also explore DSPy’s integration with 
 MLFlow for observability\, tracking\, and deployment. Attendees will leave
  with a roadmap to build scalable\, testable\, and production grade LLM ap
 plications.\n\n# Takeaways\n* Learn how DSPy’s abstractions (Signatures\
 , Modules) enable structured\, maintainable LLM applications.\n  \n* Disco
 ver how to scale LLM workflows with optimization\, multimodality\, async s
 upport\, and observability tools.\n\n\n# Audiences\n* Engineers and archit
 ects building LLM powered features.\n  \n* AI/ML practitioners exploring o
 pen source frameworks for production readiness.\n\n* Technical leaders see
 king structured approaches to integrate AI into the SDLC.\n\n# About Me\nI
  am Irfan Basha S\, a software Architect with over 15 years of experience 
 building scalable\, production grade systems at TCS\, Pramati Technologies
 \, UKG\, TechConative\, and SymphonyAI. I am passionate about AI and emerg
 ing technologies\, and I actively work with DSPy\, an open source framewor
 k\, to help teams bridge the gap between prototypes and resilient\, produc
 tion ready AI systems. I am also the Cofounder of NGI Academy\, where I fo
 cus on building a local AI community in my hometown Madanapalle by giving 
 in-person talks\, mentoring\, and encouraging others to take part in shapi
 ng the future of AI.
GEO:17.422;78.339
LAST-MODIFIED:20251028T065246Z
LOCATION:Fusion Space - Thoughtworks Technologies (India) Pvt Ltd\nHyderab
 ad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
URL:https://hasgeek.com/fifthelephant/osai-hyd-meetup/schedule/building-pr
 oduction-grade-llm-applications-with-dspy-from-simple-start-to-scalable-ex
 cellence-6QtnmbdmErb6y2PqssmtRv
BEGIN:VALARM
ACTION:display
DESCRIPTION:DSPy: building scalable\, production-ready LLM applications in
  Fusion Space in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:SchemaForge: AI-driven data workflows made easy
DTSTART:20251101T091500Z
DTEND:20251101T095000Z
DTSTAMP:20260419T202815Z
UID:session/MhXwCCinhdB37nfv48GHre@hasgeek.com
SEQUENCE:10
CATEGORIES:25-35 min talk
CREATED:20251009T042323Z
DESCRIPTION:## Abstract\nAI is moving beyond code completion to act as a t
 rue co-developer across the software development lifecycle (SDLC). From st
 ructuring raw data into validated schemas to guiding reliable data workflo
 ws\, AI can reduce friction in everyday developer and data science tasks.\
 n\nThis talk showcases **SchemaForge**\, an open-source experiment that de
 monstrates this shift:\n\n- **SchemaForge:** Goes beyond schema inference 
 by creating ready-to-use DBT models\, test rules\, and ER diagrams. It als
 o supports Python-based ETL pipelines\, making schema outputs directly usa
 ble in production workflows.\n\nThrough live demos\, we’ll explore how A
 I can ground generative outputs in structure\, validation\, and execution
 —and what this means for the future of AI-assisted development.\n\n## Wh
 at the audience will take away\n- How AI can generate DBT-ready rules from
  raw data\, with optional Python ETL ingestion support.  \n- Patterns for 
 integrating generative AI into real-world SDLC stages beyond “autocomple
 te.”  \n- Lessons from building open-source AI tools: trade-offs\, relia
 bility concerns\, and guardrails.\n\n## Format\n- **Duration:** 30 minutes
   \n- **Type:** Experiential talk with live demos  \n- **Structure:**  \n 
  - Framing: AI in SDLC beyond code completion (5 min)  \n  - Demo: SchemaF
 orge — schema inference → DBT rules → ingestion → execution (20 mi
 n)  \n  - Reflections and Q&A (5 min)  \n\n## Target Audience\n- Data engi
 neers and analytics developers who deal with repetitive schema\, ingestion
 \, and pipeline tasks.  \n- Data scientists who want faster\, structured w
 ays to prepare and validate datasets.  \n- Developers interested in buildi
 ng or contributing to AI-assisted open-source tools.\n\n## Speaker Bio\nI
 ’m Prudhvi Krovvidi\, a Data Scientist at Gramener\, where I explore how
  AI can simplify and accelerate data workflows. Most of my work ends up as
  open-source experiments on GitHub — from schema inference and quality c
 hecks to decision tree generation and AI-assisted analytics. I enjoy build
 ing lightweight tools that bring AI into everyday developer and data scien
 ce tasks — and when I’m not doing that\, you’ll probably find me out
  on my bike.\n\n- **Company:** Gramener\, Hyderabad  \n- **GitHub:** [gith
 ub.com/prudhvi1709](https://github.com/prudhvi1709)  \n- **Email:** kprudh
 vi71@gmail.com  \n- **LinkedIn:** [Prudhvi Krovvidi](https://www.linkedin.
 com/in/prudhvi-krovvidi)\n\n## Links\n- [SchemaForge](https://prudhvi1709.
 github.io/schemaforge/)\n
GEO:17.422;78.339
LAST-MODIFIED:20251028T065248Z
LOCATION:Fusion Space - Thoughtworks Technologies (India) Pvt Ltd\nHyderab
 ad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
URL:https://hasgeek.com/fifthelephant/osai-hyd-meetup/schedule/ai-as-your-
 co-developer-automating-schemas-quality-checks-ingestion-MhXwCCinhdB37nfv4
 8GHre
BEGIN:VALARM
ACTION:display
DESCRIPTION:SchemaForge: AI-driven data workflows made easy in Fusion Spac
 e in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Birds of Feather (BOF) session: Path to production
DTSTART:20251101T095000Z
DTEND:20251101T110000Z
DTSTAMP:20260419T202815Z
UID:session/ALtwdii3kTrHjK27Cgdnr8@hasgeek.com
SEQUENCE:15
CREATED:20251009T042606Z
GEO:17.422;78.339
LAST-MODIFIED:20251103T100313Z
LOCATION:Fusion Space - Thoughtworks Technologies (India) Pvt Ltd\nHyderab
 ad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
BEGIN:VALARM
ACTION:display
DESCRIPTION:Birds of Feather (BOF) session: Path to production in Fusion S
 pace in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Lightning talks
DTSTART:20251101T110000Z
DTEND:20251101T111500Z
DTSTAMP:20260419T202815Z
UID:session/UuUKg4VPntU47ZCbwhvJQz@hasgeek.com
SEQUENCE:15
CREATED:20251009T042435Z
GEO:17.422;78.339
LAST-MODIFIED:20251103T100323Z
LOCATION:Fusion Space - Thoughtworks Technologies (India) Pvt Ltd\nHyderab
 ad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
BEGIN:VALARM
ACTION:display
DESCRIPTION:Lightning talks in Fusion Space in 5 minutes
TRIGGER:-PT5M
END:VALARM
END:VEVENT
BEGIN:VEVENT
SUMMARY:Wrap-up\, feedback\, closing notes
DTSTART:20251101T111500Z
DTEND:20251101T113000Z
DTSTAMP:20260419T202815Z
UID:session/RFrcNBCCejyZJuQqDX2sMG@hasgeek.com
SEQUENCE:12
CREATED:20251009T042643Z
GEO:17.422;78.339
LAST-MODIFIED:20251103T100327Z
LOCATION:Fusion Space - Thoughtworks Technologies (India) Pvt Ltd\nHyderab
 ad\nIN
ORGANIZER;CN="The Fifth Elephant":MAILTO:no-reply@hasgeek.com
BEGIN:VALARM
ACTION:display
DESCRIPTION:Wrap-up\, feedback\, closing notes in Fusion Space in 5 minute
 s
TRIGGER:-PT5M
END:VALARM
END:VEVENT
END:VCALENDAR