90% of SDRs will be replaced by AI next year

AI Product Launches & Updates

• Qwen-Image-2512 release: Alibaba Qwen @Alibaba_Qwen announced integration into AI-Toolkit and availability on Replicate, expanding image model capabilities.

• Gemini billing update: Logan Kilpatrick @OfficialLoganK announced a streamlined billing rollout for Gemini API + AI Studio starting Jan 21 to simplify usage.

AI Tools & Applications

• Coding agent codebase analysis: Pawel Huryn @PawelHuryn highlighted that coding agents excel at documenting existing codebases for improved maintainability.

• Coding agents’ CLI abstraction: Guillermo Rauch @rauchg emphasized that the CLI is the core interface for coding agents, enabling direct OS-level actions.

Product Management Insights & Strategies

• 'Plan on a Page' checklist: George from 🕹prodmgmt.world @nurijanian shared a checklist by John Cutlefish to streamline product plans and align teams.

• '5-Minute Truth Technique': George from 🕹prodmgmt.world @nurijanian introduced this framework to avoid false positives in discovery by focusing on concise, behavior-based interviews.

• Asking input, not permission: George from 🕹prodmgmt.world @nurijanian described escaping approval hell by leading with clarity and seeking input instead of green lights.

AI Industry Developments & News

• Sales team powered by AI agents: Lenny Rachitsky @lennysan reported a shift from 10 human SDRs to 1.2 humans plus 20 AI agents delivering the same revenue, signaling mid-tier SDR roles going extinct.

• Transformer residual path improvements: Sebastian Raschka @rasbt shared tweaks to residual connections in transformers, boosting training stability beyond traditional attention and normalization optimizations.

Product Management Insights & Strategies

Random thoughts from a day of vibe coding by Peter Yang explores how AI is reshaping developer workflows. Key takeaways for PMs include:

• AI-driven planning now dominates coding sessions, making clear instructions and context management—core PM skills—even more critical.
• Interacting with AI via the terminal (CLI) is increasingly accessible, which may reduce the need for specialized GUIs or IDE plugins.
• Future interfaces will center on conversational agents—potentially voice-driven—streamlining sub-agent complexity for end users.

Call it the AI-hype-industrialcomplex by Tal Raviv challenges the community to move beyond FOMO-driven posts. He urges PMs and operators to:

• Share simple, real-world AI experiments—no need for groundbreaking demos or flashy diagrams.
• Keep it personal and practical, detailing what worked (or didn’t) so others can replicate and learn.
• Foster LinkedIn as a crowdsourced lab notebook, highlighting operator-level insights over hype.

AI Industry Developments & News

Here’s what we actually spent on AI engineering by Claire Vo breaks down last year’s AI engineering budget. Highlights:

• Significant investment continues in AI talent, signaling robust demand for engineers who can bridge product and ML development.
• Despite high spend, early-career AI roles may be under-compensated, spotlighting opportunities for PMs to rethink talent strategy and retention.

AI Product Launches & Updates

Sandbox at Vercel by Guillermo Rauch reframes coding agents around the CLI as the core abstraction. Key points:

• Agents operate at the OS level—running programs, scanning file systems, automating workflows—so CLI remains the universal interface.
• vercel.com/sandbox is introduced as an “infinite compute layer” for agents, enabling PMs to orchestrate large-scale tasks like QA, feature builds, and security audits across cloud or local machines.