Today’s issue of Hungry Minds is sponsored by:
Happy Monday! ☀️
Almost the last Monday of the year! I wish everyone happy holidays and will share soon more about the newsletter and how it’s going to evolve 🔥
Welcome to the 126 new hungry minds who have joined us since last Monday!
If you aren't subscribed yet, join smart, curious, and hungry folks by subscribing here.
📚 Software Engineering Articles
Learn the retry pattern for distributed systems resilience
Meta's root cause analysis platform scales incident management
Complete WhatsApp system design breakdown for interviews
18 practical ways engineers can leverage AI today
SSD-optimized databases maximize storage performance
🗞️ Tech and AI Trends
OpenAI launches GPT-5.2-Codex with enhanced coding capabilities
Google unveils Gemini 3 Flash, their fastest model yet
AI agents are disrupting traditional SaaS products
👨🏻💻 Coding Tip
Use torch.compile(dynamic=True) to optimize PyTorch models with variable input shapes
Time-to-digest: 5 minutes
Big thanks to our partners for keeping this newsletter free.
If you have a second, clicking the ad below helps us a ton—and who knows, you might find something you love. 💚
Voice AI is moving from experimentation to enterprise deployment.
In this new handbook, ElevenLabs breaks down…
The architectures
Trade-offs
and real-world case studies
Behind large-scale conversational agent implementations, helping leaders choose the right path to quality, compliance, and speed.
Download Building Voice Conversational Agents at Scale and see how leading enterprises are making AI conversations real.
Netflix faced significant challenges with their deployment system Spinnaker, experiencing a 4% failure rate due to transient issues. By adopting Temporal, a durable execution platform, they revolutionized their cloud operations and achieved near-perfect reliability.
The challenge: Transform a complex, failure-prone deployment system without disrupting Netflix's global operations while maintaining backward compatibility.
Implementation highlights:
Durable workflows: Migrated cloud operations to Temporal workflows that persist state and auto-recover from failures
Child workflow pattern: Implemented a two-tier workflow system for operation type resolution and execution
Gradual migration: Used feature flags to control the rollout on a per-operation basis
Activity-based design: Broke down cloud provider operations into idempotent, retryable activities
Two-hour retry window: Implemented extended retry windows to handle transient failures gracefully
Results and learnings:
Massive reliability gain: Reduced deployment failures from 4% to 0.0001%
Simplified architecture: Eliminated custom orchestration code and complex retry logic
Enhanced observability: Gained better debugging capabilities through Temporal's UI and metrics
This migration shows how modern workflow engines can dramatically improve reliability. Sometimes the best way to handle failures is to let someone else handle them for you.
ESSENTIAL (stage-fright-fighter)
How to Start With Public Speaking as an Engineer or Engineering Leader
ARTICLE (typescript-takes-the-crown)
The Trends #9: TypeScript just became the most used programming language on GitHub
ARTICLE (meta-secrets-revealed)
The Pulse #157: Internal dev tooling at Meta & the "trajectories" feature
ESSENTIAL (dev-crystal-ball)
2025 Stack Overflow Developer Survey
ARTICLE (ai-buddy-builder)
Build an AI Assistant with LangGraph, Next.js, and Auth0 Connected Accounts
ARTICLE (robots-write-code)
The State of AI Coding 2025
ARTICLE (future-cool-kids-stack)
The Vibe Coding Stack for 2026
ESSENTIAL (ai-code-whisperer)
How to Use AI Effectively in Large Codebases
Want to reach 200,000+ engineers?
Let’s work together! Whether it’s your product, service, or event, we’d love to help you connect with this awesome community.
Brief: OpenAI releases GPT-5.2-Codex, an advanced AI coding model with improved cybersecurity capabilities, featuring better long-context understanding and Windows compatibility, while implementing careful deployment strategies to balance innovation with safety concerns.
Brief: AWS unveils frontier AI agents and Nova 2 models at massive 60k-attendee conference, while CTO Werner Vogels reassures that AI will transform rather than replace developer jobs, emphasizing five key skills for future success.
Brief: As AI becomes better at writing proof scripts, formal verification of software could shift from requiring 20 person-years of work to becoming automated, potentially transforming how we ensure code reliability and reducing the need for human review.
Brief: Google unveils Gemini 3 Flash, a new AI model offering frontier-level intelligence with 3x faster speed than previous versions at lower cost, now available through Gemini app, Search, and various developer platforms.
Brief: Big tech companies knowingly sacrifice code quality by having engineers frequently switch teams, resulting in most code changes being made by people who are relatively new to the codebase, while experienced developers are overwhelmed with review duties.
This week’s coding challenge:
This week’s tip:
PyTorch's torch.compile() with dynamic=True can optimize models with variable input shapes while maintaining flexibility. The dynamic compilation tracks shape variations and creates specialized kernels for common patterns.
Wen?
Variable sequence length transformers: NLP models processing batches with different text lengths without padding overhead
Dynamic computer vision: Object detection models handling images of varying resolutions in the same batch
Reinforcement learning: Policy networks adapting to environments with changing observation spaces
Show up even when you don't want to show up.
Steve Harvey
That’s it for today! ☀️
Enjoyed this issue? Send it to your friends here to sign up, or share it on Twitter!
If you want to submit a section to the newsletter or tell us what you think about today’s issue, reply to this email or DM me on Twitter! 🐦
Thanks for spending part of your Monday morning with Hungry Minds.
See you in a week — Alex.
Icons by Icons8.
*I may earn a commission if you get a subscription through the links marked with “aff.” (at no extra cost to you).