Joe Stein

Trivy Hack Spreads Infostealer via Docker, Triggers Worm and Kubernetes Wiper thehackernews.com/2026/03/trivy-…

@raymatos I plan to give a talk about how we do auditing with Kafka for the systems in regulated environments in the near future with MCP. If interested let me know will post in the Apache Kafka NYC meetup.

@raymatos Enterprise platform with github.com/IBM/mcp-contex… and some secret sauce and SAP, Service Now, Atlassian, Microsoft365 all MCP. We are launching our WorkHQ orchestration platform for agents that will use this substrate. We build first for us then our customers. #ZeroTrust security.

Not dead yet

@jpschroeder I like it to SOAP, it’s not going away but yes something new should emerge that isn’t json bloat like we got rid of (in some industries) xml bloat. Once models are trained on TOON it’s a new ball game.

MCP ≈ Macromedia Flash It played a role, but we need to move on.

@rezoundous Try this github.com/rtk-ai/rtk

I have Codex $20 and Claude $100 plan. Somehow I hit limits on Claude regularly and never on the Codex...

@josefbuettgen There a data company, that is their moat. There is already OpenBB for years now.

the urge to build this in a weekend with claude code + opus 4.6

@VadimStrizheus It’s all converging. Human On the Loop martinfowler.com/articles/explo… this is the way

Anthropic jus dropped Code Review in Claude Code!! You can now have an entire team of agents that hunt for bugs. Vibe-coding just got even easier.

IBM in 1979:

38 researchers red-teamed AI agents for 2 weeks. Here's what broke. (Agents of Chaos, Feb 2026) AI Security arxiv.org/abs/2602.20021

Jido Concepts: Sensors Sensors connect the external world to your agents. They are a GenServer that lives at the boundary of your Agent to normalize external events into Agent Signals

@BharukaShraddha ./.claude/rules is useful too.

Most people treat CLAUDE.md like a prompt file. That’s the mistake. If you want Claude Code to feel like a senior engineer living inside your repo, your project needs structure. Claude needs 4 things at all times: • the why → what the system does • the map → where things live • the rules → what’s allowed / not allowed • the workflows → how work gets done I call this: The Anatomy of a Claude Code Project 👇 ━━━━━━━━━━━━━━━ 1️⃣ CLAUDE.md = Repo Memory (keep it short) This is the north star file. Not a knowledge dump. Just: • Purpose (WHY) • Repo map (WHAT) • Rules + commands (HOW) If it gets too long, the model starts missing important context. ━━━━━━━━━━━━━━━ 2️⃣ .claude/skills/ = Reusable Expert Modes Stop rewriting instructions. Turn common workflows into skills: • code review checklist • refactor playbook • release procedure • debugging flow Result: Consistency across sessions and teammates. ━━━━━━━━━━━━━━━ 3️⃣ .claude/hooks/ = Guardrails Models forget. Hooks don’t. Use them for things that must be deterministic: • run formatter after edits • run tests on core changes • block unsafe directories (auth, billing, migrations) ━━━━━━━━━━━━━━━ 4️⃣ docs/ = Progressive Context Don’t bloat prompts. Claude just needs to know where truth lives: • architecture overview • ADRs (engineering decisions) • operational runbooks ━━━━━━━━━━━━━━━ 5️⃣ Local CLAUDE.md for risky modules Put small files near sharp edges: src/auth/CLAUDE.md src/persistence/CLAUDE.md infra/CLAUDE.md Now Claude sees the gotchas exactly when it works there. ━━━━━━━━━━━━━━━ Prompting is temporary. Structure is permanent. When your repo is organized this way, Claude stops behaving like a chatbot… …and starts acting like a project-native engineer.

Jido 2.0: Elixir agent framework built on BEAM primitives. Isolated state, message passing, supervision, fault recovery. While Python agent frameworks reinvent the actor model with duct-taped asyncio, Erlang/OTP solved this in 1986. @mikehostetler jido.run/blog/jido-2-0-…

@mikehostetler SOAP

History will view MCP as a bloated mistake Kind of like Java Applets

Jido will grow to be the best Agent Framework in the world I won’t settle for second place jido.run

OpenAI built a cool agentic project using Elixir! 🔥 #ElixirLang

@prajdabre Keep it simple or use Apache Spark

Technical interview question: Suppose you have 5 TB worth of text data and you want to count the total number of words, how will you do this?

@matvelloso The hard part is you don’t have CICS in the cloud. It’s not the COBOL.

The market panicking because AI can rewrite COBOL ignores two things: 1-Tools that rewrite COBOL have existed for decades 2-That's not the hard part

Employees at Google and OpenAI support Anthropic’s Pentagon stand in open letter techcrunch.com/2026/02/27/emp…

@drewlevinn That’s against the law