Co-Founder & CEO @ AI Makerspace (AIM)
DL @ NVIDIA | Co-Founder & CTO @ AIM

10 people enrolled last week.
Are you a backend engineer?
Have you decided to go all in on building production AI applications?
If so, you're in the right place.
In 2026, engineers must learn how to prototype and productionize agents.
You should vibe code, yes, but keep it to frontends, docs, and boilerplate.
--
Welcome to Cohort 9, where two years later, we're tripling down on Agents.
Here's how we spend 10 weeks:
🕴️Agentic RAG (2 wks)
🧰 Complex Agents (2 wks)
📊 Evals (1.5 wks)
🚀 Certification Challenge (1.5 wks)
🚢 Production Deployments (2 wks)
🤩 Demo Day (1 wk)
Check the v0.9 curriculum for full details not available on Maven.
🏫 Our course:
Constantly evolves. Check the curriculum changelog.
Culminates in Demo Day.
Has a difficult certification process.
Is taught live, never "flipped"
Ensures you build a new project every class
Has breakouts guided by experts
Provides personalized feedback on every HW
--
AI Makerspace is on a mission to create the world's leading community for people who want to build 🏗️, ship🚢, and share 🚀 production LLM applications.
Get a flavor for our teaching style every week on YouTube.
We're all in on OSS and sharing our work; check our GitHub, LinkedIn, or profiles of any former student.
Wield AI to build AI that creates ROI. Then, forget about cash/job security issues, and live your life! #unautomatable
While we don't need to know how the engine (the LLM) works, we need to know where to go and how to put gas in the tank (the context window)
Prompt engineering, RAG, agents, multi-agent systems, and fine-tuning are the fundamental AI-specific patterns we need to manage context
Once we're operating, it's all about removing and enhancing the right context at the right step in our agentic systems
It's more important to go deep on one set of best-practice tools, understand the fundamental concepts and code, and be able to switch later
We always choose the best OSS tools on the market, according to our own opinionated views
Trust that you'll be able to switch to (enter orchestration framework, cloud service provider, LLM hosting tool, OSS model, vector DB, etc.)
Builders are great at building, but when they start sharing, they unlock the next level
Become the expert in the room during discussions about building production LLM applications
Learn to build industry-relevant $15k POCs in ~25 hours (+/- 15 hours)
Software engineers and developers who want to build, deploy, operate, and improve LLM apps in production environments.
The very best ML engineers who are pseudo-software engineers and have not lived in notebooks, and who want to build, deploy, ... etc.
Live sessions
Learn directly from "Dr. Greg" Loughnane & Chris "The Wiz 🪄" Alexiuk in a real-time, interactive format.
Lifetime access
Go back to course content and recordings whenever you need to.
Community of peers
Stay accountable and share insights with like-minded professionals.
Certificate of completion
Share your new skills with your employer or on LinkedIn.
Maven Guarantee
This course is backed by the Maven Guarantee. Students are eligible for a full refund up until the halfway point of the course.
23 live sessions • 56 lessons • 19 projects
Jan
14
Session 1: ✨ Vibe Check
Jan
16
Session 2: 🗃️ Dense Vector Retrieval
Jan
21
Session 3: 🔁 The Agent Loop
Jan
23
Session 4: 🕴️ Agentic RAG
%2520(1).png&w=768&q=75)
Not just prompt engineering, but also RAG! Learn why this fundamentally makes intuitive sense for Generative AI
Enhancing search and retrieval is the primary job of agents with access to tools (and MCP servers!) today
How tools like Deep Research offer insights into using parallelization to further scale context optimization
Live sessions
4 hrs / week
We hold class every Tuesday and Thursday at 7-9 PM ET! It's important to show up live to these sessions, as you'll always get your questions answered and get ~35-50 minutes of live breakout room time to do your homework!
Wed, Jan 14
12:00 AM—2:00 AM (UTC)
Fri, Jan 16
12:00 AM—2:00 AM (UTC)
Wed, Jan 21
12:00 AM—2:00 AM (UTC)
Projects
5-10 hrs / week
Weekly projects can be time-consuming, especially early on in the cohort, and even with the time spent live in class working on them. On average, you should budget about this much time. During Week 6, our Certification Challenge typically requires students 20-30 hours. During Weeks 9-10, Demo Day can also be quite time-consuming.
Instructor and Peer Supporter Office Hours
1-12 hrs / week
Each week, we will have our staff of 10 peer supporters (all Certified AI Engineers from the Bootcamp) and both instructors each hold an office hour. You can attend none or all of them, as you wish!
I want to thank Dr. Greg, The Wiz, and the entire AIM team. None of my projects like PromptLab (https://promptlab.so) and OpenEvals (https://open-evals.com) would have been possible without the course and the Build, Ship, and Share mentality. Simply sharing what I’ve built on X helped the framework gain visibility from thought leaders in the Vercel AI SDK and Typescript developer communities, and opened doors for new opportunities. Taking this class has truly changed my life.

Can Temizyurek
I joined AI Makerspace while looking to start my company, Publicus, and it’s also where I met my co-founder Joe. What amazed me was how inclusive the community is technical folks, researchers, and even non-developers and yet everyone still built, shipped, and shared real projects. The program completely reshaped how I use open-source tools and ship to production fast. I still rely on the teachings from Dr. Greg and Wiz today. RAGAS and EvolInstruct were game-changing takeaways that I now use to reduce hallucinations and build more reliable AI systems. It’s genuinely the best community for people who want to build real things in the real world

Akash Shetty
The AI Engineering Bootcamp gave me practical skills I use every day building Publicus. Every week, we implemented real AI packages and frameworks that I now use in production. Beyond the technical skills, the bootcamp helped me understand how to actually build agentic products—not just talk about them. Most importantly, I met my co-founder Akash there, and the network I built continues to be incredibly valuable. If you want to go from understanding AI concepts to shipping AI products, this is it.
🌐 https://publicus.ai/

Joe Noss
The Aithon journey started out of AIE4 cohort in the AI Makerspace bootcamp. A year later, the project idea has morphed into an AI-native GTM platform for regulated financial services. We are helping large Fintechs grow and expand their presence in enterprise financial services accounts. We combine external public and proprietary signals with customer’s first party data to spot whitespace and drive expansion across financial services enterprise accounts.

Nitin Gupta
I want to thank Dr. Greg, The Wiz, and the entire AI Makerspace community for creating an environment where theory becomes real traction. The program gave me the clarity and technical rigor to architect Elysia—Informa’s enterprise GenAI platform now used by 20,000+ colleagues across the globe. The “build, ship, share” mindset pushed me beyond demos into production-grade retrieval pipelines, multi-agent orchestration, and scalable AI foundations that drive measurable impact. This community has been transformative for my work, and I’m grateful for the support and inspiration throughout the journey.

Arthi Kasturirangan

Garret G.
.jpg&w=1536&q=75)
Read more at https://thedataguy.pro/blog/2025/05/build-ship-share/

We ask students to self-assess along 9 dimensions when they start and finish the bootcamp!
.jpg&w=1536&q=75)
Learn to build, ship, and share production-grade prototypes worth real cash with real customers.

We break it down heading into 2026: https://bit.ly/state-of-aie-25 (live streamed on Nov 12, 2025)

Get head of the course: https://bit.ly/langchain-v1 (live streamed on Nov 19, 2025)
https://github.com/AI-Maker-Space/LLM-Engineering-Foundations-to-SLMs-Open-Source
https://github.com/AI-Maker-Space/LLM-Ops-Cohort-1
$3,999
USD
4 cohorts