
Agentic AI builder. Loud opinions. Useful tools. 62k on LinkedIn • 41k on YouTube • Collabs DM or [email protected]
Chicago, IL
Joined July 2023
- Tweets19.2K
- Following892
- Followers35.2K
- Likes13.8K
Pinned Tweet
10 GITHUB REPOS TO BUILD YOUR OWN AI AGENTS
AI agents can do much more than answer questions.
They can remember past interactions, use tools, browse the web, collaborate with other agents, and automate entire workflows.
But you don't have to build everything from scratch.
Here are 10 open-source projects to help you get started:
1. Agno
Build AI agents with tools, knowledge, and memory.
github.com/agno-agi/agno
2. CrewAI
Create teams of AI agents that work together on complex tasks.
github.com/crewAIInc/crewAI
3. Pydantic AI
Build structured, reliable, and type-safe AI applications.
github.com/pydantic/pydantic…
4. OpenAI Agents SDK
Create agent workflows with tools, handoffs, and guardrails.
github.com/openai/openai-age…
5. Semantic Kernel
Connect AI capabilities with existing applications and services.
github.com/microsoft/semanti…
6. Letta
Build agents with persistent memory that can learn from interactions.
github.com/letta-ai/letta
7. Qwen-Agent
Develop agents with tool calling, code execution, and RAG capabilities.
github.com/QwenLM/Qwen-Agent
8. n8n
Automate workflows by connecting AI models with hundreds of app integrations.
github.com/n8n-io/n8n
9. LiteLLM
Access 100+ LLM providers through one unified API.
github.com/BerriAI/litellm
10. Oxylabs Web API MCP
Give AI agents access to live web search results and website data.
github.com/oxylabs/web-api-m…
What can you build with these?
→ Personal research assistants
→ Autonomous workflows
→ Multi-agent systems
→ AI-powered business automation
→ Agents with persistent memory
You don't need to build every component from scratch.
Start with the right framework, experiment, and build something useful.
🔖 Bookmark this list for your next AI agent project.
New users: up to ANT stock based on initial deposit and holding period, up to $200 in trading coupons with a $100 deposit, and 8.1% APY Cash Sweep for 60 days (3.6% base + 4.5% booster, enrollment required). Terms apply.
Sponsored content. Creator is paid by Moomoo Financial Inc. (MFI) and is not affiliated with MFI. Securities offered through MFI, Member FINRA/SIPC. Terms and Conditions apply. Free shares awarded under this promotion are not IPO shares; they are purchased on the open market after trading begins, at a market price that may differ significantly from the IPO offering price. Full disclosures: invest.us.moomoo.com/_disclo…
🤝 Paid partnership
128
Your AI agent isn't dumb. You're just feeding it a stale snapshot of the internet. User error. You have no-one to blame but yourself, really.
Most "agent failures" I see come down to garbage context. The model's fine. The pipeline that's supposed to fetch live pages is held together with rotating proxies, a headless browser that crashes at 2am, and a parser that breaks every time a site ships a redesign.
ScrapingBee takes that whole mess off your plate. It renders the JavaScript, gets past the anti-bot walls, retries what fails, and hands back clean Markdown or JSON your LLM can drop straight into context.
The CLI makes RAG work almost boring. Point it at a docs site, crawl it, and export chunked Markdown or NDJSON without writing glue code. It also plugs into MCP, n8n, Make, and Zapier if your agents live there.
You only pay for requests that succeed. Failed ones cost nothing.
4,700+ developers already run production pipelines on it.
Grab 1,000 free credits: scrapingbee.com/?utm_source=…
🤝 Paid partnership
Anthropic just released a 4-hour course to getting a $500k AI engineering job:
00:15 - The right way to prompt Claude
33:21 - What makes Claude act dumber on your code
01:33:39 - How Anthropic use Claude every day
02:50:56 - The fix that makes Claude way smarter This
4-hour Anthropic free course replaces about 10 paid engineering courses.
Watch it today, then read the step-by-step guide on building loops below.
10 OPEN-SOURCE GITHUB REPOS TO MAKE YOUR AI APPS MORE SECURE AND TRUSTWORTHY 🔐
Building an AI app is exciting. But making sure it behaves safely, protects user data, and gives reliable answers is a whole different challenge.
These 10 open-source projects can help you build more secure and dependable AI applications:
1. Garak
Scan LLMs for security vulnerabilities and potential weaknesses.
github.com/NVIDIA/garak
2. LLM Guard
Protect LLM applications from prompt injection, data leaks, and unsafe inputs.
github.com/protectai/llm-gua…
3. Guardrails AI
Validate model outputs and enforce structured responses.
github.com/guardrails-ai/gua…
4. Oxylabs Web API MCP
Connect AI agents to web data through an MCP server for research and data extraction.
github.com/oxylabs/web-api-m…
5. PromptBench
Benchmark LLMs across prompting techniques and adversarial attacks.
github.com/microsoft/promptb…
6. Giskard
Test LLM applications for quality, reliability, and security issues.
github.com/Giskard-AI/giskar…
7. LLM Attacks
Explore and reproduce published adversarial attacks on language models.
github.com/llm-attacks/llm-a…
8. Outlines
Generate structured and constrained outputs from language models.
github.com/dottxt-ai/outline…
9. Instructor
Get validated, structured data from LLM responses using Python and Pydantic.
github.com/instructor-ai/ins…
10. Haystack
Build LLM applications with modular pipelines and retrieval tools.
github.com/deepset-ai/haysta…
Why explore these repos?
They cover important parts of building trustworthy AI:
→ Security testing
→ Prompt injection protection
→ Output validation
→ Adversarial testing
→ Reliable structured responses
→ Better LLM application pipelines
AI apps shouldn't just be powerful. They should be safe, reliable, and predictable too.
🔖 Bookmark this list for your next AI project.
Which repository would you add to this list?
Robotics has a data problem, and so far the industry's answer has been to strap more hardware onto humans.
GoPros. UMI grippers. Teleop rigs. Exoskeletons. Every one of them drives up cost and slows the operator down. You don't get to internet-scale data that way.
Vbot just took a completely different swing. Their new model, Exo2EgoPolicy, takes ordinary third-person video (the stuff YouTube and TikTok are drowning in) and generates the first-person view a robot actually needs to learn from. It doesn't need depth maps or camera calibration. It just needs the video.
Then the Vbot Echo pipeline takes over. It reconstructs the scene, estimates hand and object poses, and retargets the human motion onto a specific robot's joints. A physics sim layer adds contact forces and torques, which you can't see in a video but a robot needs to grip anything without crushing it.
The capture side is almost comically simple. With the Echo App, you prop your phone near the work area and hit record. The footage uploads, runs through the pipeline, and comes out the other end as robot training data.
Can generated egocentric data hold up in production robotics? That's still an open question. Still, betting on volume over expensive rigs looks like the smart play to me, and Echo is now fully open for developers and research teams.
Watch the clip to see it in action.
Full breakdown from The Silicon Review: thesiliconreview.com/2026/11…
Most AI agent setups die in the config stage. Bot tokens, gateway settings, an @ on every single message.
Switch Buddy skips all of it. You install the Console, a personal assistant named after you (mine's switch-buddy-leo) asks what your team needs, then spins up the agents and drops them into a room.
In this clip it builds a whole dev team in one conversation: PM, engineer, code reviewer, bug fixer. Each one introduces itself.
They work in Jira, GitHub and Datadog as you, and you pick read or write access per agent. Try it and let me know what you think!
#ad
🤝 Paid partnership
I build integrations for clients, and my AI employee just found $20,000 of that work I never invoiced.
This wasn't padding or scope creep. It was real work, shipped and sitting in my commit history, with no line item anywhere.
Here's how it leaks. A webhook stops firing, so I fix it. A client wants one more field mapped, so I map it. A call runs long, and I tell myself it's fine.
None of it ever felt big enough to bill. Add it up and it's $20,000.
So I gave Viktor the job: read my commits and client threads, line them up against every invoice I've sent, and tell me what's missing.
He came back with 17 items across 7 clients. Each one had its proof attached: the commit, the thread, the date.
Then he drafted the invoice lines and waited. Nothing went to a client until I said yes.
I went through every line, kept the favors I meant as favors, and sent the rest.
That's the receipt, and it's the best return on anything I've bought this year.
Most consultants I know are leaving money in their git log. We just never had someone whose job it was to go look.
If you want one on your team, hire @Viktor_com. Free to start.
#AIemployee
Paid Partnership
🤝 Paid partnership
Try it at ref.viktor.com/lrodman-x-18
1
147
GOOGLE JUST DROPPED SOMETHING NOBODY SAW COMING.
An 8-minute tutorial that shows you how to build your very first AI agents from scratch.
10 GITHUB REPOS THAT GIVE YOUR AI AGENT ACCESS TO THE REAL INTERNET 🌐
An AI agent can generate answers. But what happens when it can actually search the web, explore websites, and collect information on its own?
That's where things get interesting.
Here are 10 open-source tools that can help your AI agents interact with the web:
1. Web Scout MCP
Connect AI assistants to web discovery workflows.
github.com/pinkpixel-dev/web…
2. MiniMax Search
Find relevant information across the internet.
github.com/MiniMax-AI/minima…
3. WET
Work with web data in your AI workflows.
github.com/n24q02m/wet
4. Firecrawl Web Agent
Perform web research and agent-driven browsing.
github.com/firecrawl/web-age…
5. QZ Browser Agent
Automate browser interactions for AI-powered tasks.
github.com/alibaba/qz-browse…
6. Master Fetch
Fetch content from online sources.
github.com/dondai1234/master…
7. Open WebSearch
Bring open-source web search into AI workflows.
github.com/Aas-ee/open-webSe…
8. Kagi MCP
Connect AI workflows with Kagi Search.
github.com/kagisearch/kagimc…
9. Web Search MCP
Give AI assistants access to web search capabilities.
github.com/itme-brain/web-se…
10. Oxylabs Web API MCP
Access web search and scraped data through MCP.
github.com/oxylabs/web-api-m…
What can you build with these?
→ AI research assistants
→ Automated web browsing agents
→ Real-time information gathering workflows
→ Website data extraction tools
→ AI-powered research pipelines
The real power of AI agents isn't just generating text.
It's being able to find information, interact with websites, and turn what they discover into useful work.
🔖 Save this list for your next AI agent project.
“All code is garbage except mine.” Lauren Tan, GrokBot engineer at SpaceXSI.
At the AI Engineer Code Summit, she revealed the development setup behind her team's incredible productivity.
Her approach is different from how most developers use AI agents.
Every agent mistake becomes a linting rule, not another prompt.
Agents verify their own work using videos, screenshots, and automated tests.
The result?
3,000 PRs shipped to production in a single month.
The codebase is now so clean that agents can merge most of their own PRs.
The biggest takeaway?
Don't keep fixing the same mistakes through prompts. Build systems that make those mistakes harder to repeat.
Watch this before automating your development workflow.
HOLY SHOOT. Your AI agent can do so much more when it has access to all you..
Here are 3 things August can handle:
→ Find in-network doctors near me and show me what each visit would cost.
→ Call doctors' offices, check availability, and book appointments.
→ Remind me when my next checkup is due.
I got to explore these features as part of the August beta program.
🤝 Paid partnership
We raised $12M from @Accel and @PeakXV to put AI through the bullshit American healthcare puts you through.
Navigating healthcare is hard and all of us have a list of healthcare tasks we’ve been putting off. From visiting a doctor to filing an insurance claim, the tasks pile up and get put off.
We’ve already helped 10M people understand their health. August now talks to the health system on your behalf. Tell it what you need, and August takes care of it.
In the last few weeks, August spent 24,286 minutes on hold while finding the earliest available in-network doctor appointments for our early users.
It spent 366 minutes on hold for one person alone. It also helped another save $1,255 on an insurance claim. Naturally, the ones who tried it say that they can't ever imagine going back.
I’ve held the phone for hours calling doctors, while a disease made it hard to even hold the phone. I’ve felt lost and overwhelmed when dealing with a disease and the system.
It’s why I started building August, so that all of us have a companion for our health.
8 GITHUB REPOS TO MAKE YOUR AI VIDEO CREATION WORKFLOW BETTER 🎬
Want to create better AI videos without figuring everything out from scratch?
These 8 GitHub repositories offer video examples, reusable prompts, animation styles, and creative workflows worth exploring.
1. Awesome Opus Videos
Discover AI-generated video examples and the prompts behind them.
github.com/yihui-dev/awesome…
2. Lemo Opuscar
Explore short-film creation with Claude Code across 51 visual styles.
github.com/lemomo-ai/lemo-op…
3. pdoom-video
Explore a locally rendered 4K60 karaoke music-video project.
github.com/mexicat/pdoom-vid…
4. PDoomVideo
Study another implementation of the same music-video concept.
github.com/JohnHeibel/PDoomV…
5. mg-styles-15
Explore 15 motion graphics styles with examples, prompts, and source code.
github.com/Vincentwei1021/mg…
6. ClaudeAnimationBase
Experiment with hand-drawn cartoon animations and 31 character expressions.
github.com/JohnHeibel/Claude…
7. Awesome Claude Videos
Find video-generation examples and workflows built around Claude.
github.com/athemeroy/awesome…
8. Opus Video Prompts
Explore Chinese-language AI video examples and reusable prompt ideas.
github.com/joeseesun/opus-vi…
Here's a simple way to learn from them:
→ Find a video that catches your eye.
→ Study how its prompt is written.
→ Explore the project and source code.
→ Experiment with your own creative direction.
You don't always need to start with a blank page. Sometimes, exploring what others have built is the best way to find your next creative idea.
🔖 Bookmark these 8 repos for your next AI video project!
Whenever I doubt myself, I come back to this...
Elon Musk in 2008, on the edge of losing everything, still standing in front of people and asking them to believe in an idea that the world wasn't ready for.
10 GitHub repos to level up your cybersecurity skills
Want to learn cybersecurity through real-world tools instead of just theory?
Start exploring these open-source projects:
1. Velociraptor — Hunt threats across endpoints.
github.com/Velocidex/velocir…
2. osquery — Turn machines into queryable security data.
github.com/osquery/osquery
3. Sigma — Write detection rules for security tools.
github.com/SigmaHQ/sigma
4. YARA — Identify suspicious files using patterns.
github.com/VirusTotal/yara
5. Zeek — Monitor and analyze network activity.
github.com/zeek/zeek
6. Suricata — Detect threats in network traffic.
github.com/OISF/suricata
7. MITRE ATT&CK — Learn how real-world attackers operate.
github.com/mitre-attack/atta…
8. TheHive — Manage security investigations and incidents.
github.com/TheHive-Project/T…
9. MISP — Share and analyze threat intelligence.
github.com/MISP/MISP
10. SpiderFoot — Automate OSINT and reconnaissance.
github.com/smicallef/spiderf…
The best way to build cybersecurity skills?
Explore the tools. Understand how they work. Practice in safe, authorized environments.
Don't just study security. Learn by doing.
🔖 Bookmark this list for your next learning session.
This is f*cking insane.
Anthropic CEO Dario Amodei was asked what skills people should focus on as AI starts taking over more workplace tasks.
His message is becoming harder to ignore:
AI isn't just a productivity tool anymore.
It's learning to do the actual work.
Writing code, analyzing financial reports, building presentations, researching markets, handling customer support, and managing entire workflows.
And the scary part?
It's getting better every month.
Think about what this means for someone starting their career in 2026.
A company that once needed 5 junior employees might soon need just 1 person managing AI agents.
The question is no longer whether AI will change your career.
It's whether you'll know how to work with it.
If I were starting my career from scratch in 2026, I'd learn these 14 free resources:
1. Learn the foundations
→ Google AI Essentials: grow.google/ai-essentials/
→ Microsoft Generative AI for Beginners:
github.com/microsoft/generat…
→ Elements of AI: elementsofai.com/
2. Master prompting
→ OpenAI Prompt Engineering Guide:
platform.openai.com/docs/gui…
→ DAIR AI Prompt Engineering Guide:
github.com/dair-ai/Prompt-En…
→ Anthropic Prompt Engineering Tutorial:
github.com/anthropics/prompt…
3. Learn to build AI agents
→ Hugging Face Agents Course:
github.com/huggingface/agent…
→ Microsoft AI Agents for Beginners:
github.com/microsoft/ai-agen…
→ CrewAI: github.com/crewAIInc/crewAI
→ LangChain: github.com/langchain-ai/lang…
4. Understand how AI actually works
→ Andrej Karpathy: Neural Networks Zero to Hero:
github.com/karpathy/nn-zero-…
→ Google's Machine Learning Crash Course:
developers.google.com/machin…
5. Build real-world projects
→ Awesome LLM Apps:
github.com/Shubhamsaboo/awes…
→ Anthropic Cookbook:
github.com/anthropics/anthro…
Total cost: $0.
But here's what most people will get wrong:
They'll spend the next year collecting AI tools, watching tutorials, and saving resources.
Very few will actually build something.
Don't just learn how to prompt AI.
Learn how to:
• Break complex problems into smaller tasks.
• Build workflows that save hours.
• Connect different AI tools.
• Create and manage AI agents.
• Verify AI-generated results.
• Use AI to solve real business problems.
The future won't belong only to people who know how to use AI.
It will belong to people who know how to turn AI into results.
Don't spend 2026 watching the AI revolution happen. Learn how to build with it.
Save this and send it to someone starting their career.
18 GITHUB REPOS THAT TURN YOUR AI CODING AGENT INTO AN ANIMATION STUDIO 🎬
Imagine describing an animation, and your AI agent helps build the scenes, timelines, motion effects, 3D objects, and final render.
These 18 open-source GitHub repositories can help make that workflow possible.
01. Remotion Skills — Agent guidance for video creation
github.com/remotion-dev/skil…
02. Motion Canvas — Create animated explainers
github.com/motion-canvas/mot…
03. Motion — Build smooth layout transitions
github.com/motiondivision/mo…
04. Ola — Smooth value interpolation
github.com/franciscop/ola
05. Anime.js — Create timelines and staggered animations
github.com/juliangarnier/ani…
06. Manim — Bring mathematical concepts to life
github.com/ManimCommunity/ma…
07. Shifty — Control tween playback and easing
github.com/jeremyckahn/shift…
08. Two.js — Create animated 2D vector scenes
github.com/jonobr1/two.js
09. Lottie Web — Play vector animations
github.com/airbnb/lottie-web
10. TweenJS — Sequence property animations
github.com/CreateJS/TweenJS
11. LUME — Build 3D scenes with HTML
github.com/lume/lume
12. gifenc — Encode animated GIFs
github.com/mattdesl/gifenc
13. Konva — Create interactive canvas animations
github.com/konvajs/konva
14. Blender — Work with rigging and character animation
github.com/blender/blender
15. Three.js — Build animated 3D scenes
github.com/mrdoob/three.js
16. Godot — Explore powerful game animation systems
github.com/godotengine/godot
17. Babylon.js — Create 3D animations and render scenes
github.com/BabylonJS/Babylon…
18. Proton — Add particle emitters and effects
github.com/drawcall/Proton
Think about what you can build with these:
→ Animated product demos
→ Explainer videos
→ 2D and 3D motion graphics
→ Interactive visual experiences
→ Character animations
→ Particle effects and transitions
The interesting shift is that your coding agent can become part of the creative process, not just the coding process.
Brief it. Build it. Animate it. Render it.
🔖 Bookmark this list and explore these projects!








