Stefania Druga

Unlock Multimodal AI: Intro to Interactive App Ideas

MIT research group

Google

Microsoft

Research Scientist @Google DeepMind

Understand what practical multimodal AI means and see simple examples of apps using vision, voice, and sensors data.

Build Your First Multimodal AI App

Get a peek at how accessible AI tools and APIs can help you start building your own interactive experiences.

Create Interactive Multimodal AI Apps

See a showcase of multimodal AI demo ideas (from household helpers to creative tools) to ignite your imagination.

How to Design Multimodal AI Experiences

NOTE 1: The price shown on this page is in _USD_.NOTE 2: There are not many live sessions in this bootcamp and the majority of your interaction with us will be through 1 on 1 calls with the teaching staff (5 of us).NOTE 3: Check out the incentive program on this page for ways to save money. 🟢 WHAT WILL I BE ABLE TO BUILD IN THIS COURSE?You’ll build a functional application powered by LLM agents to automate a workflow of your choice. Along the way, you'll learn to design agent workflows, and integrate required tools. By the end of the course, you'll have a working MVP ready to showcase in a public demo. You can see this playlist for examples of what previous students built: https://tinyurl.com/llm-agent-demos 🟢 HOW WILL YOU HELP ME BUILD APPS LIKE THESE IN JUST 7 WEEKS?This course is all about learning by doing. You do a real project, applying concepts you learn about AI agents with our expert guidance. Specifically:👨🏽‍🏫 1:1 sessions with the instructors to help you think through bigger picture questions👨‍🎓 Async and sync sessions with a dedicated TA for constant support, code reviews, and real-time feedback We also provide the "agents playbook" a step by step guide for building agentic systems with lots of hints, tips, and resources. 🟢 OK! THAT DOES SOUND FUN, BUT HOW DO I KNOW IF I HAVE THE PRE-REQUISITES AND I'M READY FOR BUILDING?That's the most common question we get! The best way to assess is to check out the "agents playbook" we provide for free and get a sense of what you need to do, and also check out the resources we curated. Here: https://maven.com/p/b0756d/agents-playbook 🟢 I SEE! SO, I'LL DO MY PREP BEFOREHAND, THEN WHAT SHOULD I EXPECT TO HAPPEN DURING?Think about it like this (since we're developing a product here, eh?): This is a 3-sprint product dev + 1 week of prep for the public demo.🖌️ SPRINT 1 - DESIGN (weeks 1 &amp; 2)🔨 SPRINT 2 - DEVELOP (weeks 3 &amp; 4)🚢 SPRINT 3 - DEPLOY (weeks 5 &amp; 6)🎭 PUBLIC DEMO (week 7)In each sprint, you communicate your progress or blockers on Slack, you collaborate on your designs / code on GitHub, you show up to provide weekly project updates and get feedback, and finally you do an end of sprint demo. 🟢 OK! HOW WOULD I KNOW IF I'M MOVING IN THE RIGHT DIRECTION?You will get feedback async via Slack and other platforms we use, and also you join weekly group office hours, and when you need you can book the teaching staff for a 20 min 1:1 call. In addition, each sprint will wrap up with a "sprint demo" where you can show what you've done up to that point, and your peers and the teaching staff will provide feedback. We also will invite guest experts for each of the demos to provide additional feedback:🖌️ SPRINT 1 - Sr. AI Product Experts🔨 SPRINT 2 - Sr. ML Engineers🚢 SPRINT 3 - Sr. MLOps Engineers🎭 PUBLIC DEMO - Investors, Experienced Founders, Corporate Directors 🟢 WHAT IF I DON'T HAVE A PROJECT IDEA?We will have a default project (most probably related to Deep Research) that you can participate in. If your goal is to better internalize the concepts rather than working on a specific idea, you can join that project. We commonly see that people re-take the course in the future to further develop specific ideas they have, and focus on building the foundations the first time they take the course. 🟢 COOL! I'M EXCITED TO GET STARTED, AND A DISCOUNT WOULD REALLY SWEETEN THE DEAL!Glad to hear that! There are a few options:🧑🏿‍🤝‍🧑🏼 You can find a buddy or colleague and sign up together. (20% off 2-9 seats&nbsp;|&nbsp;25% off 10+ seats)✴️ We often share discount codes with communities (including AISC members - reach out to community@ai.science if you're not part of our slack). So, check in with the communities you are part of.📆 We regularly hold events and info sessions where we share flash sales discounts valid for a day or two. Join the mailing list on this page to receive notifications about those.👨🏽‍🏫 You can apply to our Inventive Program that provides up to 75% refunds (not discounts). Apply here: https://forms.gle/YnvsoREuEX35DAxv8

Mentorship Program to Help You Build an Agent-based MVP

Course syllabus

Controlling the behavior of your agents often requires coding to include various tools, memory, or orchestration.

Python Coding

If you don't want to code, you can use no-code agent studios. We will provide specific resources but general familiarity is useful.

No-code App Building

If you are planning to code (or vibe code), familiarity with one of the coding assistants (GitHub copilot, cursor, etc), is helpful.

Coding co-pilots

We will have guest lecturers (optional) on Fridays, and project office hours on Sundays.

Fridays & Sundays

You can book time with the teaching staff to have 1 on 1 conversation about your project once or twice a week

1 on 1 Mentorship Sessions

You demo your progress every other week as you build up towards preparing for the final public demo with invited guests (experienced founders, corporate directors, and investors)

Sprints + Public Demo

Course schedule

You want to build an agentic MVP to demonstrate your skills for career progression purposes.

You want to build an agentic MVP to test the feasibility of your startup idea.

You want to build an agentic MVP to explore the new feature you want to add to your existing product

This course is for you if...

At the end of this course you will ...

Grasp how multiple LLMs can collaborate with each other, get feedback from or assign tasks to humans, ad use tools to tackle complex problems

Understand the core principles of multi-agent LLM systems

Bridge the gap by crafting strategies to seamlessly integrate LLMs with existing products, boosting user experience, productivity, and business metrics.

Integrate LLMs with existing product offerings to enhance user and customer experience

Learn what it takes to build robust multi-agent LLMs. Understand security best practices and ensure system reliability.

Understand key considerations for securing, robust, and reliable LLM systems

Design, develop, deploy, and demo an LLM agent application in 7 weeks!

Cohort-2505

Amir Feizpour, PhD

10 yrs in NLP, Founder of AISC - Community of 5k AI devs and founders

Abhimanyu Anand

Cohort-2501

Cohort-2507

Introducing a revolutionary approach to market understanding: Stop trying to validate what you want to sell. Learn to truly listen to what the market wants you to build. Imagine having direct access to the unfiltered thoughts, desires, and needs of billions of potential customers. Not through traditional market research or focus groups, but through what I call the Billion-Person Focus Group® - a revolutionary AI-powered approach to understanding what people really want. Why This Course Is DifferentMost market research courses teach you how to validate your ideas. This course teaches you something far more powerful: how to let the market guide your decisions before you even build. Using my Oracle and Architect framework, you'll learn to:→ Move beyond traditional "validation" to achieve true Product-Market Flow→ Build AI thought partners that give you deep market insights 24/7→ Tap into the authentic voice of your market at scale→ Understand not just what people do, but why they do it What Makes This Approach Unique:→ Goes beyond surface-level data to uncover emotional and cultural drivers→ Teaches you to build your own AI thought partners, not just use tools→ Focuses on deep listening rather than validation→ Combines human-centered research with cutting-edge AI technology Real Results From My Methodology:→ Helped companies pivot features to drive 40% surge in user adoption→ Reduced churn rates by 30% through deeper customer understanding→ Increased revenue by 20% by identifying hidden market opportunities→ Secured 25% market share in new markets within 6 months→ Saved millions in wasted resources and customer acquisition costs The secret?→ Stop building what you think people want and telling them what to buy.→ Instead, start listening to what the market actually needs and is telling you what to sell. If you are ready transform your vision into market-validated success then please join us. .

Ready to stop guessing and start knowing what your market truly wants?

If your events are recurring and at the same time, it might be easiest to use a single line item to communicate your course schedule to students

Tuesdays & Thursdays

Feel free to type out dates as your title as a way to communicate information about specific live sessions or other events.

May 7, 2022

Schedule items can also be used to convey commitments outside of specific time slots (like weekly projects or daily office hours).

Weekly projects

Visionary EntrepreneursBuild with confidence, not guesswork. Use AI-powered deep listening to create products your market actually wants.

Marketers &amp; Growth LeadersSpot hidden opportunities and create campaigns that resonate by understanding what truly drives your audience.

Consultants &amp; Agencies Transform client outcomes using AI-powered market intelligence. Make decisions backed by deep insights.

Solopreneurs &amp; Side-Hustlers Launch products that connect by truly understanding your niche. Stop guessing, start knowing your market.

Product &amp; Innovation Leaders Create products people love by understanding their deeper needs. Turn market insights into winning features.

This Course Is Perfect For:

What This Course Will Help You Do

Learn to use AI to test and refine your ideas before heavy investment, drastically reducing the risk of launching products that don't resonate with your audience. Say goodbye to guesswork and hello to data-driven clarity and certainty.

Validate Ideas with Confidence and Eliminate Guesswork

Dive deep into the vast world of online data to extract meaningful insights others miss. Master the Billion-Person Focus Group® technique to spot emerging trends, underserved segments and understand authentic, real-time consumer preferences before your competitors do.

Uncover Hidden Market Opportunities

Go beyond surface-level demographics to understand the deep motivations driving customer decisions. Create offerings that truly resonate with your target audience, fostering loyalty and driving sustainable growth. Build what your market craves, before they even know they want it.

Craft Products and Campaigns People Love

Gain deep insights into customer preferences to optimize your offerings and marketing strategies for maximum impact and resonance.

Optimise Existing Products and Campaigns

Use the power of AI to develop data-driven strategies for pricing, distribution, and marketing. Optimize your success while minimizing customer acquisition costs. Create messaging that deeply connects with your audience, setting your brand apart in even the most crowded markets.

Develop Winning Go-to-Market Strategies

Learn to monitor and interpret market shifts as they happen. Quickly adjust your strategies to seize emerging opportunities and maintain continuous product-market fit in dynamic environments. Stay ahead of the curve and never be caught off-guard by market changes again.

Pivot Your Strategies Proactively - Not Reactively

Save money, avoid costly missteps and accelerate your path to product-market fit. Optimize your resource allocation for maximum impact. Transform from a market follower into a visionary leader, armed with AI-powered foresight and data-driven confidence.

Minimise Risk, Maximise Success
 

Build AI-powered virtual personas or focus groups to test product concepts, predict market reactions, refine messaging or uncover unmet needs before launch. Build and deploy your own AI research agents to gather insights 24/7, transforming how you understand your market.

Create AI-Powered Virtual Personas and Market Validation Companions

Gain exclusive access to AI prompts, codebases, templates and frameworks that will revolutionize your approach to market research. Learn to craft laser-focused AI prompts that unlock hidden insights and validate your assumptions with unprecedented speed and accuracy.

Build Your DIY AI-Powered Market Research Toolkit
 

Connect, collaborate, and learn alongside visionary professionals. Get personalized support and expert guidance to tackle your unique challenges.

Join a Community of Market Innovators

Unlock the power of AI and the Billion-Person Focus Group® to decode what your market really wants in ways traditional research can't.

Abi Awomosu

Data and AI Whisperer.  Advisor to Founders and Fortune 500 Execs for 20+ Years

Cohort 2

Cohort 3

Common Challenges Faced by PMs in AI Projects 1- Justifying AI decisions and demonstrating ROI to stakeholders. 2- Navigating the hype versus the real capabilities of AI. 3- Communicating across teams without deep technical knowledge. 4- Dealing with inconsistent, missing, or messy data. 5- Aligning priorities and expectations between product and engineering. 6- Addressing ethical risks and bias in AI systems. 7- Keeping up with rapidly evolving tools, frameworks, and best practices. If these are your challenges lately then this course is for you. It’s a hands-on, live, interactive course designed for Engineers and technical Product Managers. Perfect for those with a strong familiarity with technical development and who enjoy building prototypes. What You'll Learn 1- Translate AI concepts into clear business value propositions. 2- Ground product roadmaps in realistic technical constraints and opportunities. 3- Communicate effectively with engineering teams—no deep ML expertise required. 4- Apply practical frameworks to align product vision with technical execution. 5- Navigate messy or incomplete data to build functional, user-centric solutions. 6- Identify and address ethical and bias-related risks early in development. 7- Select scalable AI tools that support long-term product growth. 8- Craft metrics and narratives that demonstrate AI impact to stakeholders. 9- Build prototypes with no-code tools like Lovable to quickly test and validate ideas. Course Format - Duration: 4-5 weeks- Live, instructor-led sessions with interactive discussions- Real-world case study following an AI product from discovery to deployment- Hands-on activities to apply frameworks and concepts- Access to a community of peers for collaboration and support Why This Course is Unique For the first time ever, a Product Manager and an Engineer are joining forces to teach AI product management. This unique collaboration combines product strategy with technical know-how, giving you the tools to not only understand AI’s potential but also execute successful AI projects from start to finish. Through this course, you’ll have access to a dedicated channel where you can ask questions, share challenges, and connect with peers and instructors. After each live session, you’ll receive a recording to review at your own pace, ensuring you never miss important details. You’ll walk away with a deep understanding of how to scope, prioritize, and launch AI products that drive tangible results and create value across teams. Who this course is not for: If you are someone who is expecting to do plenty of hands-on coding.

Leading Cross-Functional AI Product Development

June 21 - July 20, 2025

Product Managers new to AI or looking to deepen their AI skills to lead AI Products and collaborate effectively with AI tech teams

AI Engineers, Software Engineers and Engineering Leaders looking to collaborate effectively with Product leads to drive successful outcomes

Program, Project, Strategy, Marketing, UX Design/Research and Business Leaders seeking an understanding of how to bring AI Products to mark

Who is this course for

What you’ll get out of this course

How to set learning intentions and align on goals so you can collaborate meaningfully in the cohort. You'll leave with clarity on the case study and your learning journey.

Introductions & Case Study Kickoff

You’ll develop the skills to translate complex AI outcomes into clear business value, ensuring stakeholder alignment and securing the support you need to move projects forward.

Communicate AI product value to leadership and cross-functional teams

How to frame and prioritize AI use cases so you can discover viable, high-impact product ideas. You'll leave with a prioritized list and prototyping prompts.

Master iterating on AI products: gather feedback, improve, and scale

Learn AI with clear, jargon-free explanations from Deepa, an ML Engineer with over 8 years of experience. You’ll master foundational concepts so you can make informed decisions and speak the language of your tech team.

Learn core ML/AI Foundational concepts and know when to apply them in real product scenarios.

Learn to identify use cases where AI can drive real value and spot scenarios where it’s overkill. Armed with this clarity, you’ll make smarter decisions, avoid costly missteps, and build products that solve problems without creating new ones, like bias or user distrust.

Understand where AI can be leveraged effectively and where it won’t help

Foster true collaboration with your technical teams, ensuring your product vision aligns with their realities. You’ll gain frameworks to bridge the gap between product goals and technical constraints, enabling you to ship AI products faster and more effectively.

Collaborate seamlessly with data scientists and engineers to build winning AI products

Tackle the ethical minefield of AI with confidence by learning to identify and address issues like bias, fairness, and privacy concerns before they derail your product; keeping your users safe and your company out of hot water.

Navigate ethical and regulatory challenges in AI

Learn to evaluate the right technologies for every stage of product development. Whether it’s choosing the best LLM for a chatbot or a vectorization tool for search functionality, you’ll build a toolkit that saves time, cuts costs, and delivers results tailored to your goals.

Leverage the best AI tools to enhance productivity as part of daily workflows

You’ll learn to account for AI’s unique challenges, like data prep, model training, and iteration cycles, so you can deliver on time without compromising quality or over-promising.

Design realistic and relevant AI timelines to meet business objectives

Among the sea of Gen AI Tools, use the right ones that can drastically improve your productivity as a product manager

Leverage Gen AI Tools to increase productivity

Build higher-quality AI products! Live course for PMs & Engineers to apply proven frameworks & bridge product-tech gaps.

Deepa Krishnamurthy

Avinash Mahalingam

Cohort 1

Latency budgets, desktop permissions, SIP rules, plus the AI minefield—hallucinations, accent-biased transcriptions, token-burst rate limits, and model drift—turn “just add voice” into months of painful debugging even for senior engineers. WebRTC jitter ruins timing, loose prompts spark off-brand replies, and a single bad transcript cascades into wrong actions and angry users. Over four weeks you’ll build three voice products—desktop, browser, and telephony—side-by-side with the maintainers of Vapi, Pipecat, LiveKit, and Whisper, plus Ivan and Nicolay, who’ve shipped these systems in production. You’ll master the shared scaffolding (stream segmentation, prompt stitching, cost/latency meters) and the AI guardrails (real-time validation, confidence scoring, speculative decoding) that keep voice assistants responsive, factual, and customer-safe across every channel and model. But don't just trust us that Voice AI is here to stay and having it's "ChatGPT" moment. "Humans interact with businesses in many ways, but one way hasn't changed much in almost 100 years—and that's phone calls.Today, over a trillion calls exist between a business and a customer....new voice models and conversational LLMs are now incredibly good ... startups are ... making voice AI bots that are indistinguishable from humans." - Gustaf Alströmer, YC -- in a call for Voice AI startups "For enterprises, AI directly replaces human labor with technology. It’s cheaper, faster, more reliable — and often outperforms humans. Voice agents also allow businesses to be available to their customers 24/7 to answer questions, schedule appointments, or complete purchases...For consumers, we believe voice will be the first — and perhaps the primary — way people interact with AI." - Olivia Moore, a16z -- AI voice in consumer There are already meeting bots that talk to you over Zoom, language coaches that help you learn Spanish in a webapp, ambient assistants that sit on your laptop, listen, and help you out when you need some input. "Voice UX" will keep growing in surface: We will move into cars, AR glasses, smart speakers, and other areas we haven't even considered yet. Each channel breaks in its own way—latency on the web, permissions on desktop, SIP rules on telephony. One demo can’t teach you all of that. That's why we build three separate voice AI applications in this course. A webapp, a native (MacOS) app, and a telephony app. 1. Native macOS Meeting Assistant – records your mic locally, takes live notes, pushes tasks to Notion, and pings you in Slack before deadlines. Learn how the most successful app to date (Granola) does it. Manual note-taking in back-to-back calls burns 6 h/week and important information still slips. 2. Web-based AI Sales Coach – simulates tough customers, scores every response, and shows real-time coaching tips without breaking flow. Learn how to live update UX based on an ongoing conversation. New reps take 6 months to hit quota; live coaching is expensive.3. Telephony Booking Bot – calls clients, confirms appointments, handles DTMF/silence, and writes results straight into your CRM. Learn how to reliably call and handle diverse accents. Staff spend hours calling clients; no-show rate ~45 %. Why you care as a student- These metrics resonate with CTOs, PMs, and investors—your demo isn’t a toy.- Each channel teaches a different “gotcha”: OS sandbox, browser jitter, telephony regs. Master once, reuse forever.- Portfolio proof: three repos that shout “I can ship voice products anywhere users speak.” After that, the next interface is just more plumbing. The tools you’ll learnVapi – voice routing without IVR hellPipecat – low-latency audio transformsLiveKit – WebRTC that survives bad networksWhisper/Elevanlabs/AssemblyAI – fast, accurate transcriptionOpenAI &amp; Gemini Realtime – millisecond-level reasoning Hands-on workshops with the engineers who wrote these libraries Exclusive Access: Connect directly with the engineers building these tools through dedicated workshops and Q&amp;A sessions. Learn from those who know these technologies best.- LiveKit founder workshop: learn about WebRTC and how to make networking a breeze. - OpenAI real-time API creators: learn how to best prompt real-time model. - more workshops from speakers from ElevenLabs, AssemblyAI, Vapi, Pipecat will be added soon. Prerequisites (read this)- Comfortable in TypeScript/JavaScript (async/await, streams, React or similar)- Basic REST &amp; WebSocket chops- Familiarity with Git and command-line tooling If you’ve never shipped production code, this bootcamp will overwhelm you.

Voice AI is here, but it is freaking hard to build

Interactive workshops and implementation guidance We will determine the best time based on the students and their time zones.

Live Sessions

Get 1:1 help with your prototypes and code before you dive into the weekend. We expect that you start to get into the code after the live session and then already have questions before the weekend.

Office Hours

Your weekly implementations. These can be adjustments to existing code or complete implementations of real-time voice applications. You can bring your own idea or build upon our existing applications. We are happy to help you with both, but recommend the later.

The “Build-It-Now” CTO racing to add voice; needs production-ready blueprints, cost controls, and multi-channel code now.

Software &amp; AI Engineer – General software &amp; AI engineers exploring voice; seek hands-on repos to learn streaming audio, LLM prompts,...

By Demo Day you’ll have a macOS meeting assistant, a WebRTC sales-coach webapp, and a Twilio/Vapi booking bot running on your own account—ready to show a boss, investor, or client.

Ship 3 real voice products in 4 weeks

We hand you working repos, infra scripts, and latency / cost benchmarks. Ship voice features 2–3 × faster than starting cold.

Save 24+ engineering hours on “figuring it out”

<ul><li>Meeting assistant users cut note-cleanup by 6 h/week and miss 0 action items.</li><li>Sales coach slices rep ramp-time from 6 → 3 weeks.</li><li>Booking bot drops no-shows by 40 %, freeing staff for upsell calls.</li></ul>

Quantifiable business impact you can brag about

Live coding + AMA sessions with:<ul><li>Russel D'Sa (LiveKit) – WebRTC, LiveKit, Voice Agents</li><li>Ivan Leo &amp; Nicolay Gerold (Aisbach) – prompt stitching, production guardrails</li></ul>

Hands-on with the maintainers

Automated latency alerts, hallucination detectors, and ASR-accuracy checks you can drop straight into any future voice project—so bugs surface in CI, not in prod.

Plug-and-play test & guardrail suite

Download-ready spreadsheet + benchmarks scripts that score every major ASR (Whisper, AssemblyAI, Deepgram), TTS (ElevenLabs, Polly), routing layer (Vapi, Twilio), and realtime LLM (OpenAI, Gemini) on latency, cost, language coverage, and hallucination rate. Run npm run bench.

Voice-AI Tool Selection Playbook

Get into a members-only Slack with maintainers (Ivan, Nicolay) and other builders. Dedicated channels for #latency-bugs, #prompt-design, and #show-your-metrics guarantee you can paste logs, share PRs, book 15-min pairing slots, and get answers during the course.

Private Discord “War Room” for Real-Time Help

Voice AI is about to have it's own ChatGPT moment. Learn how to build applications that listen and act. 

Lightning Lessons

Unlock Multimodal AI: Intro to Interactive App Ideas