Why AI Agents Fall Apart After Hours of Work
Nuno Campos
Most AI agents can run for hours, but almost none can tell you if the work they did was right. Learn how Nuno Campos is solving this problem at Witan Labs.
Aug 19, 2026 · 32:04 BACK TO HOME
BUILDING
Learn about development practices, technical implementation, and building robust products.
What 20M Pull Requests Reveal About AI's Impact
Nicholas Arcolano
Jellyfish's Nicholas Arcolano breaks down 20M pull requests from 200K engineers: coding speed has doubled, so why hasn't business output caught up?
Jul 27, 2026 · 36:39Why Every Piece of Software Got 3-10x Less Valuable
Samuel Colvin
Samuel Colvin on shallow moats, slop forks, and what survives in the agentic era.
May 11, 2026 · 40mAI Transformation Without Layoffs: How Wix Did It in Under a Year
Asaf Yonay
Invest in mindset alongside tooling: how Wix scaled AI-native transformation across 5,500 employees in less than a year.
May 7, 2026 · 32mThe Cyborg Executive
Adam Ben-David
Embrace the 'Cyborg' model: augment human capabilities with AI for optimized software development outcomes
Feb 19, 2026 · 8m 24sThe Research Revolution
Adam Ben-David
Reinforce research in R&D: Dedicate teams for continuous discovery, foster an open-source model, and actively integrate AI in research.
Feb 19, 2026 · 6m 23sFrom Prompt Engineering to Context Engineering
Adam Ben-David
Harness AI's ability with context engineering and bespoke tooling for optimal software development performance
Feb 19, 2026 · 6m 33sPrivacy, Security & The Future of Federated AI
Chong Shen Ng
Balance privacy with utility in Federated AI: adopt additional privacy-enhancing technologies, manage privacy-utility trade-offs, and control model access
Feb 10, 2026 · 6m 23sBringing the Model to the Data
Chong Shen Ng
Adopt federated AI: enhance data privacy, leverage local context, and fine-tune models for industry-specific insights
Feb 10, 2026 · 8m 18sCODE REVIEW AT AI SPEED
Maciej Korolik
Integrate AI in code review: use AI for meticulous checks, automate changelogs, and separate tasks among coding agents
Jan 15, 2026 · 7m 25sDeveloper Community and Technical Architecture
Bobak Tavangar
Invest in quality data and harness AI power: dive deeper on data understanding, optimize systems for recall, and leverage AI for intelligence extraction
Jan 14, 2026 · 9m 5sHardware Software and AI Embodiment
Bobak Tavangar
Combine hardware-software approaches, integrate AI for enhancing user experiences, and develop discerning buying decisions in AI-assisted software development.
Jan 14, 2026 · 6m 31sOpen Source and the AI Coding Reality Check
Bobak Tavangar
Embrace AI discerningly: Foster deep-rooted open-source movement and ensure thorough understanding of AI-built systems
Jan 14, 2026 · 7m 22sThe Contrarian Vision Why AI Glasses Not VR Graphics
Bobak Tavangar
Leverage AI in wearable tech for real-world understanding and privacy, not graphic simulations
Jan 14, 2026 · 5m 34sFIGMA TO CODE IN MINUTES
Maciej Korolik
Optimize design to code translation: adopt AI-tools and ensure well-structured, clear design files for enhanced efficiency
Dec 17, 2025 · 7m 2sBUILD AI TOOLS, NOT SAAS
Maciej Korolik
Leverage AI-first development: Build bespoke, efficient internal tools faster and maintain them with evolving documentation and shared best practices
Dec 17, 2025 · 6m 19sHUMAN-IN-THE-LOOP BEATS VIBE CODING
Maciej Korolik
Integrate 'human-in-the-loop' planning in AI-assisted software development for quality control and code comprehension
Dec 17, 2025 · 6m 35sWHEN SPEED BREAKS SOFTWARE - FAST TEAMS, FRAGILE SYSTEMS
Zbigniew Sobiecki
Navigate the AI 'Velocity Paradox': balance rapid feature development with system complexity management
Dec 16, 2025 · 5m 29sBUILDING WITH INFINITE EXPERTS - THE AGE OF INFINITE TEAMS
Zbigniew Sobiecki
Reimagine software development: integrate AI as team members, prioritize judgment over skills, and champion exploration over convention
Dec 16, 2025 · 6m 14sDESIGNING DETERMINISM IN AI - HOW AI BECOMES PREDICTABLE
Zbigniew Sobiecki
Embrace context engineering: manage chaotic AI systems through contextual cues to drive predictable outcomes
Dec 16, 2025 · 5m 5sIS CODING ABOUT TO DIE? - HOW ARE PROGRAMMING JOBS TRANSFORMING
Krzysztof Zablocki
Harness AI to supercharge your workflow: Speed up testing, encourage language learning, and evolve team structures for efficiency.
Dec 16, 2025 · 7m 4sDESIGNING CONTROL AROUND AI - HOW TO MAKE AI CODE RELIABLE?
Krzysztof Zablocki
Master control over AI chaos: Delegate coding to AI, focus on system design, and establish deterministic checks for reliability and efficiency
Dec 16, 2025 · 6m 9sCONTEXT ENGINEERING FOR BETTER AI - CONTEXT IS THE NEW PROMPT
Krzysztof Zablocki
Harness context engineering: control AI model's context for cleaner coding and progressive task disclosure for efficiency gains.
Dec 2, 2025 · 5m 30sPrompts Are Code: The New Version Control Challenge
Oji Udezu
Operationalize prompts: version them, test for drift, and treat changes like code releases.
Sep 29, 2025 · 0m 20sAI at the Core vs AI at the Edge
Oji Udezu
Move AI from edge to core: make models the primary logic layer, with code orchestrating prompts, tools, and safeguards.
Sep 29, 2025 · 0m 29sDelegate Work to Silicon: The New Development Paradigm
Oji Udezu
Automate outcomes, not steps: hand full jobs to AI where possible so products deliver finished work rather than dashboards.
Sep 29, 2025 · 0m 29sFrom Edge to Core: The AI-Native Revolution
Oji Udezu
Rebuild around models: treat LLMs as programmable building blocks and redesign systems for probabilistic behavior.
Sep 29, 2025 · 6m 30sThe New Development Stack: Four Layers of AI Abstraction
Chris Rickard
Adopt the new stack: UserDoc for requirements → V0 for prototypes → export code → Cursor/Windsurf to finish the last 20%.
Sep 29, 2025 · 0m 23sThe Hidden Orchestration: 5,000 AI Interactions for 70 Features
Chris Rickard
Expose the hidden work: orchestrate thousands of AI calls behind simple UX, with monitoring to ensure reliability at scale.
Sep 29, 2025 · 0m 26sThe 40% Production Code Threshold: Where AI Contribution Plateaus
Chris Rickard
Expect ~40% AI‑written code in production: keep human ownership of architecture, enforce reviews and tests, and reserve complex logic for engineers.
Sep 29, 2025 · 0m 26sBuild vs Buy Becomes Meld: The Hybrid Architecture
Steve Brown
Combine buy and build: wrap third‑party models with your logic and UX, and write code only where it compounds advantage.
Sep 29, 2025 · 0m 19sFrom Databases to Digital Employees: The New Build Target
Steve Brown
Design digital employees: specify goals, permissions, tools, and escalation rules instead of database tables and forms.
Sep 29, 2025 · 0m 28sThe 10x Productivity Paradox: Build More, Not Fire People
Steve Brown
Use AI gains to out‑ship competitors: keep teams intact, raise throughput goals, and funnel the 10× lift into more experiments, features, and customer value.
Sep 29, 2025 · 0m 20sFrom Hexadecimal to Human Intent: The Final Abstraction Layer
Steve Brown
Build from human intent: move beyond code to directing digital employees, defining roles, guardrails, and desired outcomes.
Sep 29, 2025 · 3m 24s