Published 27 Jan 2026
Duration: 2424
Software development experts discuss the importance of general-purpose AI agents, scalable solutions, and simplicity in design, with a focus on expanding AI capabilities beyond coding to tackle broader software engineering challenges.
If you rely on complex scaffolding to build AI agents you aren't scaling you are coping. Thibault Sottiaux from OpenAIs Codex team joins us to explain...
The podcast explores the transformation of OpenAI's Codex into a general-purpose AI agent designed to adapt across various software development scenarios. It emphasizes the value of building scalable, foundational systems over focusing on specific products, reflecting the "bitter lesson" in AI that scalable solutions tend to prevail over domain-specific optimizations. The discussion advocates for simplicity in design to avoid unnecessary complexity and outlines a vision for the agent to evolve beyond coding into broader software engineering tasks.
Looking ahead, the team aims to develop multi-agent networks that could significantly enhance productivity. Open-sourcing the agents repository is seen as a way to encourage innovation and collaboration within the community. Insights from earlier open-source efforts and language transitions, such as moving from TypeScript to Rust, underscore the importance of adaptability and user feedback. Future challenges include improving the agent's efficiency in long-running tasks and tailoring its behavior to better match individual user preferences and workflows.
12 May 2026 Its Tuesday and your tech stack is obsolete (again). Now what? | Theory Ventures Bryan Bischof
Explores AI and data science challenges through collaborative research, hidden technical hurdles like inference optimization, educational gaps, critiques of hype cycles, frameworks like Apex, and the balance between AI's role as a driver/tool, cautioning against prioritizing speed over predictability in integration with human workflows.
8 May 2026 Goblins in prod, the messy middle of AI adoption, and everything is a harness now
AI development challenges include NFT-based identities, avatar integration, data leakage issues like "Goblin Invasion," risks of bias in retraining, agent misalignment, workforce disparities, open-source frameworks like Lattice, lightweight tools, and the need for systemic safeguards to address technical and organizational deployment hurdles.
5 May 2026 Teach the primitives or watch your competitor define them | Basetens Philip Kiely
AI education and developer advocacy are reshaping engineering through AI-native workflows, agentic skills, cross-functional collaboration, and structured integration frameworks to address scaling challenges and drive innovation in AI-centric practices.
1 May 2026 Tokemaxxing scoreboards, the vegan LLM from before 1931, and 30% of the web is now AI-generated
Evolving AI challenges include rising costs, security risks, ethical dilemmas, and the need for balanced strategies in adoption, such as managing usage-based pricing, implementing safeguards, addressing token costs, and ensuring real-world impact measurement beyond metrics, alongside concerns about semantic diversity and human-AI collaboration.
28 Apr 2026 Giving robots a brain | Intrinsics Brian Gerkey
Advancements in AI, particularly large neural networks, drive robotics from rigid automation to adaptable, real-world systems via software-defined hardware, open-source platforms like ROS, and collaborative initiatives addressing reliability, simulation integration, and modular design for democratization.