HalfStack Vienna 2026 โ€ข Nov 19, 2026

Bringing Our Platform to Life

Daniel Cranney

Turning 3,000+ hours of video into searchable data required ditching the AI hype. See how local models, automated transcription pipelines, and agent-ready Markdown solved a massive content discovery problem.

Starting at 0:01 ยท Watch from the beginning

Pause
Mute Enter Fullscreen
#1 about 6 min

Navigating developer cynicism and discovering the community platform

Because developers quickly reject inauthentic marketing pitches, building a valuable community platform requires listening directly to user needs and offering accessible resources for generalists and specialists alike.

#2 about 4 min

Testing developer skepticism with AI news headlines

Evaluating outrageous artificial intelligence news headlines reveals how easily plausible-sounding agent failures blur the line between reality and fiction for software engineers.

#3 about 5 min

Managing massive video libraries and diverse content formats

Confronting an overwhelming library of thousands of hours of conference talks, podcasts, and livestreams requires extracting high-value short-form clips without flooding users with noise.

#4 about 6 min

Tracking industry shifts toward AI guardrails and security

Moving past initial AI hype and chaos, the software industry now prioritizes practical implementation through human-in-the-loop workflows, agent orchestration, and security guardrails.

#5 about 3 min

Replacing keyword matching with context-aware semantic search

Traditional keyword search wastes developer time by lacking context, so modern content platforms must interpret full natural-language questions to surface accurate, multi-perspective answers.

#6 about 6 min

Automating video clip extraction using local transcription models

Manually scrubbing through hours of video to create social media reels and searchable platform moments is inefficient, so using local transcription models to identify timestamped highlights unifies the content pipeline.

#7 about 6 min

Balancing human user interfaces with agent-friendly markdown

While preserving well-crafted user interfaces for human developers, platforms must also convert content into markdown and provide agents.md files to integrate seamlessly into terminal and agentic workflows.

#8 about 4 min

Building interactive community features driven by user feedback

Instead of forcing unnecessary AI features into production, listening to community feedback leads to engaging interactive UI additions like a live-scoring news trivia game.

Matching moments

2:16 min

Welcome and updates from the WeAreDevelopers team

Raphael De Lio Raphael De Lio +3 ยท LIVE

1:44 min

Summarizing developer experience and artificial intelligence companions

Robert Hoffmann Robert Hoffmann +1 ยท LIVE

3:28 min

Implementing conversational developer experience with artificial intelligence

Graziano Casto ยท LIVE

1:23 min

Developer adoption of integrated AI chat features

Julia Kasper ยท Coffee With Developers

8:52 min

Chunking and repurposing video content with artificial intelligence

Simon Cox +2 ยท LIVE

1:21 min

Using artificial intelligence to reimagine developer experience

Alexander Trusheim Alexander Trusheim +1 ยท World Congress 2025

Upcoming sessions on this topic

Starts in 5 days

Conference India 2026

November 25โ€“26, 2026

Bengaluru, India