Daniel San
@dani_avila7 • 34,563 subscribers
Head of AI at https://t.co/04TQdit36I | OSS project https://t.co/pEjytZiAFd (32k⭐️)
Shorts
Videos
Introducing Jev Model Router for Claude Code This Claude Code Mod lets you use Jev through its direct TypeSafe AI API or Vercel AI Gateway With every request you send to Claude Code, Jev classifies the subagent model, main model (only at session start to avoid breaking the cache), and effort level Install it with one command: npx claude-code-templates@latest --mod productivity/jev-model-router Full Mod details:
Daniel San206,776 views • 1 month ago
Fireworks in the US just hit different I’m from Chile, where fireworks are illegal, so I’d never experienced anything like this Last night in Michigan, a neighbor invited us to watch his backyard show He’s just a guy… just one dude with $15k!!! worth of explosives and a 30 minute fireworks show in his own backyard 4th of July in the US is next level 🇺🇸
Daniel San412,620 views • 3 months ago
Introducing Jev Skill Suggestion for Claude Code Now you can keep your skills completely out of the context window until they’re actually needed Set your skills as user invocable only, and the Mod sends the available skill list to Jev For every request, Jev classifies which skill best matches the task, then injects only that skill into Claude’s context Works with both the TypeSafe AI API and Vercel AI Gateway Install it with one command: npx claude-code-templates@latest --mod productivity/jev-skill-suggestion Full implementation details on the Mod page at
Daniel San81,381 views • 1 month ago
We saw exactly this in my previous startup RAG + vector DB gives decent results, but agentic search over the repo (glob/grep/read, etc) consistently worked better on real-world codebases. We even pushed further: RAG + embeddings + AST + tree-sitter. The quality was excellent But exactly as Boris Cherny mentions: staleness and privacy, you need continuous re-indexing, and all the code and embeddings must live on your servers. In practice, fast models + bash-style agentic search ended up outperforming general RAG search, even if it requires more tool calls This is what we built 👇
Daniel San448,294 views • 8 months ago
Deepseek running locally and privately for autocompletion in VSCode! 🙌 In less than a minute, I'll show you how to download Deepseek-coder and set it as the autocompletion model in VSCode. You’ll need to use ollama to download the model and CodeGPT to select it as the autocompletion model. Enjoy the best models running locally with :)
Daniel San993,349 views • 1 year ago
Claude Code 2.1.63 dropped with a new built-in command: /simplify It reviews your changed code for three things: - Reuse opportunities (duplicated logic, extractable patterns) - Code quality (readability, naming, structure) - Efficiency (unnecessary complexity, redundant operations) Then it actually fixes what it finds. Not just suggestions, it edits. How to use it: 1. Make your code changes as usual 2. Run /simplify 3. It analyzes your diff, finds issues, and applies fixes In the video I ran it after finishing a PR review and noticed it spawned 3 parallel agents using Haiku 4.5 to do the analysis... fast and cheap Try it out 👇
Daniel San372,786 views • 7 months ago
Claude Code subagents can nest 5 levels deep now Boris Cherny announced it, and today I finally got to try it, Here's the full chain running end to end: - main - project-auditor // level 1 - structure-checker // level 2 - import-validator // level 3 - dependency-tracer // level 4 - style-sync // level 5 Each level runs in its own context window Only the top-level summary returns to main, depth 5 is the hard cap, that agent can't spawn further
Daniel San152,011 views • 4 months ago
I highly recommend making your Haiku subagents exploration-only in Claude Code You can force your subagents into exploration only mode and leave code changes to your main agent I added two new components to Claude Code Templates: 1- Subagents Explore Only (Setting) Uses SubagentStart to instruct subagents to only read, search, and report findings without modifying files 2- Subagent Read Only Guard (Hook) Uses PreToolUse to block file writes and disallowed commands. If a subagent tries to modify files, the hook blocks the action Install both: npx claude-code-templates@latest --setting hooks/subagents-explore-only --hook security/subagent-read-only-guard Your main agent keeps its normal permissions, while subagents focus on exploration
Daniel San15,320 views • 10 days ago
Introducing DeepSeek R1:1.5b Running Locally in Cursor! In less than 4 minutes, I set up the DeepSeek R1:1.5b model, download it, and run it locally to seamlessly work with code in Cursor The video is shown in real-time, with the model impressively running on an Intel Core i5 😱 using Ollama and the CodeGPT extension 👇
Daniel San584,360 views • 1 year ago
Quick demo of how I use session to session messaging in Claude Code + Ghostty (Save this and try it in your own workflows) 1- I always start with a regular session and a panel running claude agents 2- The first session becomes the orchestrator. It creates the initial context, then distributes the work across other sessions using forks 3- Each fork starts with that context + a prompt telling it what to work on 4- I rename each session with ctrl + r to keep things organized and make it easy to message them from the main session 5- Using @ I send messages to the sessions that are running and ask them to report their task status back, so the main session always knows what everyone is working on This way, I only need 2 or 3 panels open and can switch between tasks as needed The days of opening endless terminal panels and running Claude Code 100 times just to look productive are over! What we need now is ORCHESTATION one main session coordinating everything through session messaging + Agent View And the key is context, build it in the main session first, then fork from there That’s what makes the entire workflow click
Daniel San63,627 views • 2 months ago
Boris has Claude Code loops running on cron all day - One babysits his PRs and fixes CI - Another keeps CI healthy - Another pulls Twitter feedback every 30 min and clusters it His Claude Code setup is simple and pretty close to mine That's what I like about Claude Code. Hundreds of options, but the simplest setup just works, you probably don't need more than that And since one of those loops reads Twitter every 30 min, Boris Cherny (or his Claude on cron) is reading this post whether he wants to or not 😅
Daniel San121,592 views • 5 months ago
Claude Code just shipped v2.1.69 with a new built-in skill: /claude-api It detects your language, picks the right surface (direct API, tool use, or Agent SDK), and loads the relevant reference docs directly into context. Covers streaming, batches, files API, structured outputs, tool use patterns, MCP, also sets model defaults so you're not guessing. Useful when you're building on top of Claude and don't want to context-switch to the docs every 5 minutes.
Daniel San152,831 views • 7 months ago
New Claude Code Mod… it keeps getting more hackable and customizable This Mod lets you see every subagent running in your session You won’t lose track of them anymore... they stay visible in the sidebar, where you can review what each one has done and how much context it has used Install it with one command: npx claude-code-templates@latest --mod ui/agent-flow Link:
Daniel San22,324 views • 25 days ago
Jev + Vercel Sandbox Mod in Claude Code This Mod spins up a sandbox on Vercel, while Jev classifies which commands should run inside the sandbox and which ones should run locally Install: npx claude-code-templates@latest --mod security/jev-vercel-sandbox Link: Mods are a great way to customize your Claude Code workflows and connect other platforms directly into the agent loop
Daniel San16,889 views • 19 days ago
The more I use Claude Projects, the more convinced I am that this thread based workflow is the right direction One of the hardest parts of working with Claude Code is managing tasks across different sessions Worktrees help, naming sessions and using colors help. But it’s still hard to keep track of everything and visualize the history A session doesn’t just end and disappear It contains decisions, executed commands, PRs, and most importantly, a STATE: - Working - Waiting for you - Idle - Resolved Everything stays organized inside the same project, together with the Library, Pull Requests, and Routines It finally gives you one place to see what each session is doing, what happened before, and what still needs your attention My only feedback so far is that threads run inside a remote sandbox, and sometimes I need to execute commands directly on my own computer Everything else... pretty much flawless 👏
Daniel San24,189 views • 1 month ago
Claude Code Templates just hit 100+ Skills on Installing is easy - one click to copy the install command, paste it in your project's .claude folder, done. - 67K npm installs - 13K+ GitHub stars - fully open source Thanks to the community for contributing Skills and components. In the video I show how to add a Data Privacy Compliance Skill to your project.
Daniel San164,913 views • 10 months ago
Claude Code /simplify is now /code-review Now accepts an effort parameter: - low: short, scoped, latency-sensitive tasks that aren't intelligence-sensitive - medium: cost-sensitive work that can trade off some intelligence for fewer tokens - high: balances tokens and intelligence. Minimum for intelligence-sensitive work, or to cut spend vs xhigh - xhigh: best results for most coding and agentic tasks. Recommended default on Opus 4.7 - max: can improve performance on demanding tasks but shows diminishing returns and tends to overthink. Test before adopting broadly
Daniel San86,169 views • 5 months ago
