Dmitry Bakaev

CTO, Product Manager, Builder @ AI-First Enterprise

[Dmitry Bakaev @bakaevd]

@dbabbs @tryprofound @lightspeedvp @sequoia @kleinerperkins @MattEvantic @Saga_Ventures @spc Agent

[Dmitry Bakaev @bakaevd]

Parents and educators, I've been involved with Alpha School and 2 Hour Learning, revolutionizing education with AI and innovative student motivation strategies. The results are transformative.

Join our event featuring MacKenzie Price, Co-Founder of Alpha School, and William Stixrud, Ph.D., and Ned Johnson, co-authors of the bestselling book, "The Self-Driven Child: The Science and Sense of Giving Kids More Control Over Their Lives."

WHEN: September 18
WHERE: Venues across the U.S. (Live & Live Streamed)

When young people feel connected to a real sense of purpose — and have caring adults in their corner — they don’t just get through the day… they own it!

You’ll hear research, real-world stories, and practical strategies to help unlock your kid’s fire and passion without piling on stress or pressure. You’ll walk away with fresh insights to help your child grow into a capable, confident young adult.

Learn more and REGISTER:
https://t.co/91v7tKThTc

[Dmitry Bakaev @bakaevd]

Although ChatGPT 5 stole the show today, I still recommend checking out this new https://t.co/XgXMVcO9tu course ... if you are not already using Claude Code.

[Andrew Ng @AndrewYNg]

I'm thrilled to announce the definitive course on Claude Code, created with @AnthropicAI and taught by Elie Schoppik @eschoppik. If you want to use highly agentic coding - where AI works autonomously for many minutes or longer, not just completing code snippets - this is it.

Claude Code has been a game-changer for many developers (including me!), but there's real depth to using it well. This comprehensive course covers everything from fundamentals to advanced patterns.

After this short course, you'll be able to:
- Orchestrate multiple Claude subagents to work on different parts of your codebase simultaneously
- Tag Claude in GitHub issues and have it autonomously create, review, and merge pull requests
- Transform messy Jupyter notebooks into clean, production-ready dashboards
- Use MCP tools like Playwright so Claude can see what's wrong with your UI and fix it autonomously

Whether you're new to Claude Code or already using it, you'll discover powerful capabilities that can fundamentally change how you build software.

I'm very excited about what agentic coding lets everyone now do. Please take this course!

https://t.co/HGM8ArDalK

[Dmitry Bakaev @bakaevd]

@kevinlu625 We are building awesome AI apps for education at https://t.co/9tNnp2nqH7. We've been using v0, Bolt and Cursor. Would love to try this.

[Dmitry Bakaev @bakaevd]

Key takeaway from 2 days at #AgentConference: What a difference a year makes! AI agents have officially arrived. The tooling has matured significantly, and discussions have shifted from flashy demos to real production use cases that deliver value. https://t.co/VWOm2UxCb5

[Dmitry Bakaev @bakaevd]

Today I went through this https://t.co/fG15HsIIyv course that exposes the capabilities of OpenAI's o1 model. The difference in just two years from the original 3.5 and 4 models is massive. I strongly recommend these short courses from Andrew Ng's deeplearning ai to anyone who wants to stay current on the latest AI tech but has a tight time budget.

[Dmitry Bakaev @bakaevd]

@AndrewYNg amen

[Dmitry Bakaev @bakaevd]

This has been my message to the engineering team over the past year. Being a coder is going to become less valuable. Being able to dive deep into the business problem and understand WHAT needs to be built and WHY will grow in value.

[Andrew Ng @AndrewYNg]

Writing software, especially prototypes, is becoming cheaper. This will lead to increased demand for people who can decide what to build. AI Product Management has a bright future!

Software is often written by teams that comprise Product Managers (PMs), who decide what to build (such as what features to implement for what users) and Software Developers, who write the code to build the product. Economics shows that when two goods are complements — such as cars (with internal-combustion engines) and gasoline — falling prices in one leads to higher demand for the other. For example, as cars became cheaper, more people bought them, which led to increased demand for gas. Something similar will happen in software. Given a clear specification for what to build, AI is making the building itself much faster and cheaper. This will significantly increase demand for people who can come up with clear specs for valuable things to build.

This is why I’m excited about the future of Product Management, the discipline of developing and managing software products. I’m especially excited about the future of AI Product Management, the discipline of developing and managing AI software products.

Many companies have an Engineer:PM ratio of, say, 6:1. (The ratio varies widely by company and industry, and anywhere from 4:1 to 10:1 is typical.) As coding becomes more efficient, teams will need more product management work (as well as design work) as a fraction of the total workforce. Perhaps engineers will step in to do some of this work, but if it remains the purview of specialized Product Managers, then the demand for these roles will grow.

This change in the composition of software development teams is not yet moving forward at full speed. One major force slowing this shift, particularly in AI Product Management, is that Software Engineers, being technical, are understanding and embracing AI much faster than Product Managers. Even today, most companies have difficulty finding people who know how to develop products and also understand AI, and I expect this shortage to grow.

Further, AI Product Management requires a different set of skills than traditional software Product Management. It requires:
- Technical proficiency in AI. PMs need to understand what products might be technically feasible to build. They also need to understand the lifecycle of AI projects, such as data collection, building, then monitoring, and maintenance of AI models.
- Iterative development. Because AI development is much more iterative than traditional software and requires more course corrections along the way, PMs need be able to manage such a process.
- Data proficiency. AI products often learn from data, and they can be designed to generate richer forms of data than traditional software.
- Skill in managing ambiguity. Because AI’s performance is hard to predict in advance, PMs need to be comfortable with this and have tactics to manage it.
- Ongoing learning. AI technology is advancing rapidly. PMs, like everyone else who aims to make best use of the technology, need to keep up with the latest technology advances, product ideas, and how they fit into users’ lives.

Finally, AI Product Managers will need to know how to ensure that AI is implemented responsibly (for example, when we need to implement guardrails to prevent bad outcomes), and also be skilled at gathering feedback fast to keep projects moving. Increasingly, I also expect strong product managers to be able to build prototypes for themselves.

The demand for good AI Product Managers will be huge. In addition to growing AI Product Management as a discipline, perhaps some engineers will also end up doing more product management work.

The variety of valuable things we can build is nearly unlimited. What a great time to build!

[Original text: https://t.co/OIeAQXpriK ]

[Dmitry Bakaev @bakaevd]

Was shown a sample AI generated podcast by a coworker today. Super impressive! Here is more on it:

[Andrej Karpathy @karpathy]

NotebookLM is quite powerful and worth playing with
https://t.co/EMHIjc15iU

It is a bit of a re-imagination of the UIUX of working with LLMs organized around a collection of sources you upload and then refer to with queries, seeing results alongside and with citations.

But the current most new/impressive feature (that is surprisingly hidden almost as an afterthought) is the ability to generate a 2-person podcast episode based on any content you upload. For example someone took my "bitcoin from scratch" post from a long time ago:
https://t.co/7ajZNZ0BGi
and converted it to podcast, quite impressive:
https://t.co/ZZn0LJgsnu

You can podcastify *anything*. I give it train_gpt2.c (C code that trains GPT-2):
https://t.co/gDrAqix4Iv
and made a podcast about that:
https://t.co/bgcwmQr5d7
I don't know if I'd exactly agree with the framing of the conversation and the emphasis or the descriptions of layernorm and matmul etc but there's hints of greatness here and in any case it's highly entertaining.

Imo LLM capability (IQ, but also memory (context length), multimodal, etc.) is getting way ahead of the UIUX of packaging it into products. Think Code Interpreter, Claude Artifacts, Cursor/Replit, NotebookLM, etc. I expect (and look forward to) a lot more and different paradigms of interaction than just chat.

That's what I think is ultimately so compelling about the 2-person podcast format as a UIUX exploration. It lifts two major "barriers to enjoyment" of LLMs. 1 Chat is hard. You don't know what to say or ask. In the 2-person podcast format, the question asking is also delegated to an AI so you get a lot more chill experience instead of being a synchronous constraint in the generating process. 2 Reading is hard and it's much easier to just lean back and listen.

[Dmitry Bakaev @bakaevd]

"The most common mistake of smart engineers is to optimize the thing that should not exist"

https://t.co/3yWis35jfi

43:47 - 48:44

[Lex Fridman @lexfridman]

Here's my 8.5 hour conversation with @elonmusk & @neuralink team.

It is the longest podcast I've ever done! 🤯 It's a fascinating, super technical, and wide-ranging conversation.

It's here on X & everywhere else, including YouTube: https://t.co/0kYbFNj2Su

On X, it's split up into 3 parts (in replies to this post) due to the current 4 hour limit for video. I continue to push the X engineering team to increase the limit... up it to 10 hours let's go! 🤣

Thank you to Elon Musk, DJ Seo, Matthew MacDougall, and Bliss Chapman who are part of the amazing Neuralink team: @elonmusk, @djseo_, @matthewmacdoug4, and @chapman_bliss.

And thank you to Noland Arbaugh (@ModdedQuad) who is the first human to have a Neuralink device implanted in his brain.

Timestamps:
0:00 - Introduction
0:49 - Elon Musk
4:06 - Telepathy
10:45 - Power of human mind
15:12 - Future of Neuralink
20:27 - Ayahuasca
29:57 - Merging with AI
34:44 - xAI
36:57 - Optimus
43:47 - Elon's approach to problem-solving
1:01:23 - History and geopolitics
1:05:53 - Lessons of history
1:10:12 - Collapse of empires
1:17:55 - Time
1:20:37 - Aliens and curiosity
1:28:12 - DJ Seo
1:36:20 - Neural dust
1:43:03 - History of brain–computer interface
1:51:07 - Biophysics of neural interfaces
2:01:36 - How Neuralink works
2:07:26 - Lex with Neuralink implant
2:27:24 - Digital telepathy
2:38:27 - Retracted threads
2:44:01 - Vertical integration
2:50:55 - Safety
3:00:50 - Upgrades
3:09:53 - Future capabilities
3:39:09 - Matthew MacDougall
3:44:58 - Neuroscience
3:52:07 - Neurosurgery
4:03:11 - Neuralink surgery
4:22:20 - Brain surgery details
4:38:03 - Implanting Neuralink on self
4:53:57 - Life and death
5:03:17 - Consciousness
5:06:11 - Bliss Chapman
5:19:27 - Neural signal
5:26:19 - Latency
5:30:59 - Neuralink app
5:35:40 - Intention vs action
5:46:54 - Calibration
5:56:26 - Webgrid
6:19:28 - Neural decoder
6:40:03 - Future improvements
6:48:59 - Noland Arbaugh
6:49:08 - Becoming paralyzed
7:02:43 - First Neuralink human participant
7:06:45 - Day of surgery
7:24:31 - Moving mouse with brain
7:49:50 - Webgrid
7:57:52 - Retracted threads
8:06:16 - App improvements
8:13:01 - Gaming
8:23:59 - Future Neuralink capabilities
8:26:55 - Controlling Optimus robot
8:31:16 - God
8:33:21 - Hope

[Dmitry Bakaev @bakaevd]

There is absolutely no reason to throw GenAI/Agents at a problem that can be deterministically solved via RPA.

[Dmitry Bakaev @bakaevd]

Came across this read on O'Reilly that's largely reflective of our 1.5 yrs of working with LLMs:
https://t.co/H1xIZ86mMa

[Dmitry Bakaev @bakaevd]

Anybody who's built RAG systems on corporate data will appreciate this. We've gone through our struggles over the past 2 years, with lots of lessons learned. Hopefully this will make your life simpler if you are just starting out.

[Andrew Ng @AndrewYNg]

Data preprocessing is critical for building effective RAG systems. Our new short course, Preprocessing Unstructured Data for LLM Applications, taught by @mrobinson0623 of @UnstructuredIO, demonstrates important but sometimes overlooked aspects of RAG systems:

- How to extract and normalize content from diverse formats like PDF, Powerpoint, and HTML to expand your LLM's knowledge
- Enriching data with metadata to enable more powerful retrieval and reasoning
- Applying document layout analysis and vision transforms to process embedded images and tables

Then you’ll use all these skills and build a RAG bot that draws from a corpus that includes PDF, PowerPoint, and Markdown documents.

Please sign up here: https://t.co/AM3rmZJmNF

[Dmitry Bakaev @bakaevd]

Zero to one

[Elon Musk @elonmusk]

Precisely

[Dmitry Bakaev @bakaevd]

Better results can be achieved by using AI Agents than more powerful models with simple prompting. A good (and short) update from Andrew Ng on the latest in the AI agentic workflows: https://t.co/UcpvN9Ob00

[Dmitry Bakaev @bakaevd]

So I looked through NVDA's AI Foundation models and the AI courses they recently published. All meant to to create lock-in within their ecosystem. They've taken an early lead in the AI race and are getting outsized profits. Where there are outsized profits there is competition. It will be an interesting business case to watch re what NVDA will do to defend their position over the next N years.

https://t.co/v4DHYrjt96

https://t.co/ssW9g7BlWG

[Dmitry Bakaev @bakaevd]

@svpino We use Deepgram and love it.

[Andrew Ng @AndrewYNg]

Our new short course, Knowledge Graphs for RAG, is now available! Knowledge graphs are a data structure that is great at capturing complex relationships between data of multiple types. By enabling more sophisticated retrieval of text than similarity search alone, knowledge graphs can improve the context you pass to the LLM and the performance of your RAG applications.

In this course, taught by @akollegger of @neo4j, you’ll
- Explore how knowledge graphs work by building a graph of public financial documents from scratch
- Learn to write queries that retrieve text and data from the graph and use it to enhance the context you pass to an LLM chatbot
- Combine a knowledge graph with a question-answer chain to build better RAG-powered chat systems

Sign up here! https://t.co/N3gceKrvib