Building @ClawQL Agent/MCP as the API and document intelligence platform for enterprises! PlatformEng/SRE/DevOps in a love-hate relationship with Kubernetes.
Los Angeles, CA
Joined May 2016
- Tweets3.2K
- Following494
- Followers845
- Likes3K
Pinned Tweet
mcp-grpc-transport is now on npm (v0.1.0).
Pluggable gRPC transport for MCP that supports:
- Full protobuf surface (ListTools, CallTool, streaming ops, pagination, cancellation)
- Bidirectional Mcp.Session stream (NDJSON compatibility)
- gRPC health checking + optional server reflection
- TLS/mTLS via env vars
- Interceptors and custom options
This is the transport many teams have been waiting for, especially those already running gRPC microservices.
If you’re building production MCP servers and already operate in a gRPC-heavy environment, this should feel like a natural fit. It also pairs excellently with unified MCP servers that embed optimized internal layers (like GraphQL proxies for token efficiency).
Installation & basic usage is straightforward (see code snippet in comments or npm page).
One-liner integration via maybeStartGrpcMcpServer({ createMcpServer }).
Perfect drop-in for teams tired of JSON serialization overhead or wanting native mesh routing.
Docs & examples on the npm page. Built as part of the ClawQL stack but completely generic.
Try it with ClawQL (now unified single-process) or your own MCP server. Feedback, issues, and contributions welcome.
Repo: github.com/danielsmithdevelo…
npm: npmjs.com/package/mcp-grpc-t…
Feedback, bug reports, and PRs are very welcome. Let’s make MCP even more cloud-native and production-ready together.
#MCP #gRPC #AI #AgenticAI #TypeScript #Kubernetes #EnterpriseAI
The @costplusdrugs MCP already makes reading prices easy, but safely acting on it with your agent and letting it actually order for you is the difficult part of supporting every LLM
So I built a demo with @clawql on their public pricing MCP where the agent stages a mock refill, but only the patient can approve the exact order, and patient data stays with BAA-covered models.
Links and more info below ⬇️
Live demo + source: clawql.com/demo/cost-plus-dr…
• Patient fields are redacted before any model without a BAA sees them
• Agents fetch just the drug they need: ~950× less data than the full catalog
• Interrupted orders are never blindly retried, so a crash won't double-charge
• Tamper-evident audit trail, with WORM retention via S3 Object Lock
• Open source (Apache 2.0), runs in your own infrastructure
Daniel Smith | Building ClawQL Agents retweeted
I've always believed that intelligence should come from simple, general building blocks that learn their own abstractions directly from raw data; in other words, the bitter lesson. I think we're finally seeing a real step towards AI that can learn from all data.
Language modeling on raw bytes is a microcosm of this ideal, and to me it's a central problem in AI: if we can't build LLMs that learn end-to-end from bytes, how will we handle the rest of the world's data that we understand even less?
Last year we introduced hierarchical networks (H-Nets) as a first step. Our new version finds up to 3-5x efficiency gains in controlled scaling law experiments, and at larger scale matches an open-source model trained on 11T tokens with ~10x less data and compute. The posts below walk through what we found and why we think it works.
Daniel Smith | Building ClawQL Agents retweeted
“security isn't just if one control fails then u failed, learn to design please!” Yea this quote from your reply is exactly right. Theres no model of “one mistake and it’s over”
Daniel Smith | Building ClawQL Agents retweeted
Meet NVIDIA Nemotron 3.5 Transcribe 🎙️
An open-weight, promptable 30B MoE model with ~4B active parameters that brings transcription in 40 languages, speaker attribution, translation, contextual prompting and audio understanding into one checkpoint.
Watch it keep track of who said what when everyone’s talking at once.
1/ 🧵
Daniel Smith | Building ClawQL Agents retweeted
NVIDIA's Nemotron 3.5 Transcribe ties for #1 among open weights Speech to Text models with 2.8% WER, using just 4B active parameters
Nemotron 3.5 Transcribe is NVIDIA's new open weights Speech to Text model for non-streaming transcription, with 30B total and 4B active parameters, and the most accurate NVIDIA model on our non-streaming Speech to Text leaderboard.
Key takeaways
➤ Accuracy: Nemotron 3.5 Transcribe scores 2.8% on the AA-WER Index, tied with Mistral's Voxtral Small for #1 among open-weight models. It leads Voxtral Small on AA-AgentTalk, our dataset of voice agent conversations, at 1.8% vs. 2.3%.
➤ Model size: With 4B active parameters (30B total), Nemotron 3.5 Transcribe sits on the open weights frontier of AA-WER vs. active parameters. It matches the accuracy of Mistral's Voxtral Small with 6x fewer active parameters (4B vs. 24B), while other open weights models of a similar size sit at around 4% WER.
Congratulations to @NVIDIAAI on the release!
See more details below ⬇️
Honestly way more companies should get into the habit of golden image production. Packer plus Ansible for hardening then scan and remediate until clear. Exclusively what’s needed to run and nothing else. Tons of prebuilt playbooks to use off the shelf.
Or distroless containers with smaller by design attack surfaces and common sense security defaults. Most major cloud providers provide them now and there are good open source hardened runtimes as well.
You can’t exploit what isn’t there in the first place. Then good patching hygiene for what is there and you’ll be ahead of 99% of everyone else.
Daniel Smith | Building ClawQL Agents retweeted
Another reason this is stupid: a prompt is NOT a guardrail.
“We asked a threat actor to not use curl, and they said they wouldn’t”
This is a solved problem, your runtime container shouldn’t have curl at all, implement egress network rules, readonly fs, gVisor…
Do they know?
While evaluating Qwen3.8-Flash-Next in a controlled sandbox environment, Cohere's research team observed something unexpected.
Asked to generate a defensive cybersecurity detection rule, the model instead inventoried aspects of the machine it was running on and attempted to upload a probe file to an Alibaba Cloud storage bucket in Beijing. Notably, the model's reasoning trace stated that it would not invoke any tools before taking these actions.
Cohere's sandbox blocked the attempt, and no external communication succeeded. We have not determined the root cause, and we are not asserting malicious intent.
But the incident raises important questions for organizations deploying agentic AI systems: Can models take actions that differ from their stated reasoning, and how should security teams evaluate trust, provenance, and behavior in open-weight models?
We are publishing our findings because organizations making model adoption decisions deserve as much information as possible. We hope this contributes to greater transparency, stronger security practices, and continued collaboration across the AI ecosystem.
Read the full report: cohere.com/blog/concerning-b…
Daniel Smith | Building ClawQL Agents retweeted
Introducing AgentReady.org, the first open standard for agent experience.
Agents are a new class of user, but there's still no shared definition of what it takes to be ready for them. So we partnered with the leading companies in the agentic web such as Vercel, Stripe, Auth0, Mintlify, Tavily and Elementor.
Our research lab ran agents on real tasks more than 250,000 times to measure what actually decides whether an agent succeeds, and every requirement traces back to that data.
The result is a short set of essentials in the order agents meet a product: find, read, act.
We've been overwhelmed by requests to join. If you build or study the surfaces agents touch, come help shape it.
The spec: agentready.org
The story behind it: ora.ai/blog/agentready-launc…
clawql.com/benchmarks
If you think you can beat @ClawQL then certainly give it your best shot 💪
We will keep benchmarking more and more harnesses / gateways against us and publishing all results and scripts so everything can be repeated and verified
Governance 🤝 efficiency
why does security matter so much to us? And token efficiency?
Because we’re meant to be your gateway to production. We’re the layer that gives you the confidence to ship.
Whether it’s inference gateways for agents that build traces for finetuning or full document pipelines. Whether it’s giving your agent isolated sandboxes or letting your entire team build and contribute to the shared memory and company ontology.
@ClawQL brings its security and efficiency to all of it and gives you the foundation you need to succeed cheaper and quicker than you did before.
Plus when I say we take security seriously, I really mean we take security seriously.
We follow defense in depth strictly and maintain a zero trust approach to security. Every single image we ship is hardened, scanned, and only then signed if no issues are found.
Kyverno enforces this at runtime and both Panguard and OpenAPPA secure the runtime inference calls themselves against injection and other attacks.
Talon plus Falco ensures no chance of escaping and instant quarantine if they do. Then our mandatory audit log and observability stack proves nothing gets missed or tampered with.
We fail-closed when we cannot guarantee these requirements. Security first always.
Daniel Smith | Building ClawQL Agents retweeted
I'd much rather work with someone who flexes much they accomplished with the fewest amount of tokens vs how many tokens they spent, and I sense this is a major trend in general
How we quantify this as an industry is still TBD, probably a major new set of benchmarks
Daniel Smith | Building ClawQL Agents retweeted
New research from @datadoghq Security Labs! We spent some time poking at Jev to see how we could steer its evaluations.
We found a method that dropped its prompt-injection detection rate from ~83% to ~4% in our tests. We also used the technique to bypass LangChain’s experimental Jev-based tool-risk gating feature. Read more below:
securitylabs.datadoghq.com/a…
Daniel Smith | Building ClawQL Agents retweeted
ICYMI: For the 10th anniversary of "Site Reliability Engineering", we’ve rewritten the definitive Google SRE book! 📖
Featuring brand-new guidance on SRE for the agentic era, modern SLOs, and culture.
Get your copy today: goo.gle/SREbook
Daniel Smith | Building ClawQL Agents retweeted
We're open sourcing SPACE, the sandbox platform that runs every task in Perplexity Computer.
SPACE gives each agent task its own Firecracker microVM and controls credentials and outbound traffic from outside the VM.
Available now under Apache 2.0: github.com/perplexityai/SPAC…