chtefi
- Karma
- 0
- Created
- ()
- Submissions
- 0
- Testing a Kafka Proxy: Taming Millions of Permutations (conduktor.io)
- Loop Engineering: When Generation Gets Cheap, Judgment Gets Expensive (sderosiaux.substack.com)
- How to Tell If Your Kafka Self-Service Is Working? (medium.com)
- MCP is definitely not dead (sderosiaux.substack.com)
- Most agent reliability problems are data engineering problems (sderosiaux.substack.com)
- Agents Are Microservices with a Brain (blog.dataengineerthings.org)
- Do LLMs Dream? The Post-Transformers Generation (sderosiaux.substack.com)
- Is Anthropic Enshittifying their core product? (sderosiaux.substack.com)
- $200 subscription VS $3,650 in compute (sderosiaux.substack.com)
- Kafka Explorer – Configuration, Protocol, Wire Format, Errors, Flink SQL (kafka-options-explorer.conduktor.io)
- Paper Cuts #2: RAG is dead, long live memory (sderosiaux.substack.com)
- Decisions, extracted knowledge, handoff context: reasoning data infrastructure (sderosiaux.substack.com)
- How to stop your AI agent from gaming its own KPI (sderosiaux.substack.com)
- Algorithm names are the new cognitive activators. Your vocabulary is your tool (sderosiaux.substack.com)
- Claude Code told me what tools it needs to work faster (sderosiaux.substack.com)
- Semantic closure: why compilers know when they are right and LLMs do not (sderosiaux.substack.com)
- Prompt optimization can outperform reinforcement learning on LLMs (sderosiaux.substack.com)
- Internet is eating itself. What's next? Model collapse and AI data crisis (sderosiaux.substack.com)
- Matryoshka embeddings: How to make vector search 5x faster (sderosiaux.substack.com)
- 1500 Lines of Markdown vs. 15000 Lines of Python (sderosiaux.substack.com)
- The 70% AI productivity myth: why most companies aren't seeing the gains (sderosiaux.substack.com)
- Why most software startups don't need VCs anymore (most) (sderosiaux.substack.com)