AI WITH Rithesh

Production AI Agents: 15 Real Failures (and the Architecture That Fixes Them)

My aim for creating this series was to learn about real Production AI Agents failures and how we can overcome them

A demo agent that works is easy. An agent that survives production is a different animal. This series dissects 15 real, documented incidents : Replit deleting a production database,Air Canada held liable for its chatbot, MCP tool-poisoning, self-replicating agent worms and, for each one, the architecture teams actually shipped to stop it happening again.

Each episode is a self-contained 7 minute deep dive: the incident, the mechanism derived from first principles, runnable before-and-after code, a non-obvious insight, and a concrete checklist for your own system. Watch in order the ideas compound: context engineering, enforcement outside the model, provenance and trust boundaries, verification gates, observability, and knowing what not to automate.

This is for engineers, applied-AI practitioners, and technical leaders shipping agents.

www.youtube.com/playlist?list...
Do you see value in these playlists ?

What such topics do you want me to cover? Add in comments

1 week ago | [YT] | 2

AI WITH Rithesh

LLMOps → Agentic AIOps: The Hands-On Playlist (2026)

Everything you need to take large language models and AI agents from a demo to production - one focused, practical topic per video. This playlist assumes you know the basics (LLMs, prompts, RAG, agents) and goes straight to the operational layer that actually decides whether a system is reliable.

Across 11 videos you'll cover prompt management, evaluation, observability,
production RAG (RAGOps), guardrails, inference cost, and the full agentic stack :ReAct agents, multi-agent orchestration, memory, and the MCP and A2A protocols. Every video includes runnable code, the real tools you'd use, and company case studies (Klarna, Uber, LinkedIn). Every claim is tied to a named paper or a real deployment.

www.youtube.com/playlist?list...

3 weeks ago (edited) | [YT] | 5

AI WITH Rithesh

Have done hundreds of interviews for AI Engineer roles over the period of last 2 years.

People easily throw the buzzwords but dig a little deeper they struggle with their answers

Based on this experience and for my own preparation I created this course on Udemy

It collects the toughest, most commonly asked interview questions across the five pillars of modern GenAI engineering and answers each one the way a strong candidate should: derive the mechanism instead of naming it, put a number on it (bytes, recall, latency, GPU memory), and state the tradeoff you are accepting.

Everything is organized into three tiers that build on each other, Beginner foundations, Intermediate mechanisms, and Advanced/Architect-level system design, across five modules:

► Retrieval-Augmented Generation (RAG): chunking, hybrid search and Reciprocal Rank Fusion, re-rankers, "lost in the middle," knowledge conflict, and designing RAG for 500M+ documents under 200ms.
► Transformers : Tokens and embeddings, Attention, Multi-head attention, KV cache, GQA, RoPE, FlashAttention & MoE ,PagedAttention
► Embeddings & Vector Databases: the distributional hypothesis, anisotropy, HNSW, product quantization, Matryoshka embeddings, sharding to a billion vectors, drift, and zero-downtime re-indexing.
► LLM Agents & Agentic Systems: the agent loop, tools, memory design, MCP, orchestration, multi-agent systems, long-horizon autonomy, security, and observability.
► LLM Fine-Tuning: pretraining vs SFT vs alignment, loss masking, chat templates, LoRA/QLoRA/DPO, the GPU memory math, distributed training with ZeRO and FSDP, and serving many fine-tunes at once.
► LLM Evaluation: offline vs online eval, LLM-as-a-judge and its failure modes, RAG-specific metrics (faithfulness, answer relevance, context precision/recall), benchmark contamination, regression testing, and catching silent quality decay in production.

Every answer comes with background, a worked example, and real code, shown framework-agnostic and with modern SDKs where relevant. The trickier claims are checked against the original papers.

This is for anyone preparing for AI Engineer, ML Engineer, Applied Scientist, or the newer LLM/GenAI Engineer roles, and for practitioners who'd rather understand the machinery than copy recipes

www.udemy.com/course/genai-engineer-interview-prep…

This is a collection of my YT videos but you get everything in a single place and ad-free

3 weeks ago (edited) | [YT] | 4

AI WITH Rithesh

Why LLMs needed Reinforcement Learning

How did a next-word predictor become an assistant that follows instructions, refuses harm, and reasons through olympiad math? 

The answer is reinforcement learning and this series explains every landmark, from the RLHF recipe behind ChatGPT to the verifiable-reward RL behind DeepSeek-R1.

14 standalone videos. Watch any one on its own, or go in order. Every episode is built the same way: plain-English intuition first, then the math derived from
first principles, TWO simple worked examples you can follow on paper, the failure modes that bite in practice, and a head-to-head comparison with the alternatives.


What the series covers:
• The alignment gap: why pre-training alone is not enough
• Reward models, policy gradients, PPO (the clip + the KL leash)
• Cheaper paths: RLAIF/Constitutional AI, DPO, rejection sampling
• The reasoning era: process rewards, o1, GRPO, DeepSeek-R1 / RLVR
• Agentic RL and how to choose a method for your own project

www.youtube.com/playlist?list...

4 weeks ago | [YT] | 6

AI WITH Rithesh

Anthropic just figured out how LLMs work!
Anthropic's interpretability team just published something remarkable: evidence that language models keep a small, privileged set of representations they can report, steer, and reason with sitting atop a much larger sea of automatic processing. It looks a lot like the "global workspace" that neuroscientists associate with conscious access in humans.They call this J-space
Checkout my video on J-space
https://youtu.be/uTUOGQtJ-uc

1 month ago | [YT] | 4

AI WITH Rithesh

Why LLMs needed Reinforcement Learning

How did a next-word predictor become an assistant that follows instructions, refuses harm, and reasons through olympiad math? The answer is reinforcement learning and this series explains every landmark, from the RLHF recipe behind ChatGPT to the verifiable-reward RL behind DeepSeek-R1.

14 standalone videos. Watch any one on its own, or go in order. Every episode is
built the same way: plain-English intuition first, then the math derived from
first principles, TWO simple worked examples you can follow on paper, the failure
modes that bite in practice, and a head-to-head comparison with the alternatives.

No hand-waving and no LaTeX wall — every formula comes with real numbers.

What the series covers:
• The alignment gap: why pre-training alone is not enough
• Reward models, policy gradients, PPO (the clip + the KL leash)
• Cheaper paths: RLAIF/Constitutional AI, DPO, rejection sampling
• The reasoning era: process rewards, o1, GRPO, DeepSeek-R1 / RLVR
• Agentic RL and how to choose a method for your own project

www.youtube.com/playlist?list...

1 month ago | [YT] | 4

AI WITH Rithesh

The theory that AI/ML interviews at OpenAI, Anthropic, Meta, Google, Apple, Microsoft, Netflix, and Tesla actually test — one animated, narrated deep dive per question. No hand-waving: every episode derives the idea from first principles, shows a runnable code snippet, animates the intuition, and ends with exactly what the interviewer is probing for.

This playlist covers the 16 fundamentals every ML engineer is expected to own cold: the bias–variance decomposition and double descent, parametric vs non-parametric models, the curse of dimensionality, linear and logistic regression from their likelihoods, why the MLE of variance is biased, MLE vs MAP vs Bayesian inference, generative vs discriminative models, L1 vs L2 regularization, convexity and non-convex training, why gradient descent works (and fails), backprop by hand, entropy/cross-entropy/KL, bagging vs boosting, VC dimension, and universal approximation with no free lunch.

Each video is standalone with its own intro and takeaway — start anywhere, or watch in order for a full theory refresher. New clusters (Classical ML, Deep Learning, ML System Design, and more) drop regularly.

▶ Watch in order, or jump to the question you're weakest on.
🔔 Subscribe for the rest of the AI/ML Engineering Interview series.

#MachineLearning #MLInterview #MLEngineer #InterviewPrep #DataScience #DeepLearning

www.youtube.com/playlist?list...

1 month ago | [YT] | 4

AI WITH Rithesh

If you're prepping for an ML, GenAI, or backend interview, embeddings and vector databases come up first - and they sound simple until you have to explain why dense beats sparse, or what a vector database does that Elasticsearch can't.

This beginner tier opens with six BACKGROUND questions (the history and the why), then nine foundations - each with a concrete code sample (sentence-transformers, FAISS, gensim, numpy). By the end you'll be able to explain what an embedding really is, how
it's trained, why all three similarity metrics agree on normalized vectors, and what "approximate" search buys you.

What you'll learn:
- Why we moved from TF-IDF to dense embeddings, and a short history of the field
- The distributional hypothesis, static vs contextual embeddings
- What a vector database adds over an inverted index and a NumPy array
- How embeddings are trained, hard negatives, similarity metrics, ANN, evaluation

https://youtu.be/p5RbP5y-_2o

1 month ago | [YT] | 2

AI WITH Rithesh

"Strong on definitions. Light on tradeoffs."
Picture the candidate who walks out of a GenAI interview sure they crushed it. They defined RAG, embeddings, agents, fine-tuning, and LLM eval cold.
Then came the second question: "Okay, but why does that happen, and what would you do about it?"
Silence.
That second question is what quietly decides GenAI interviews right now. So I built an entire course around surviving it.
It collects the toughest, most commonly asked questions across the five pillars of modern GenAI engineering and answers each one the way a strong candidate actually should:
→ Derive the mechanism instead of naming it
→ Put a number on it: bytes, recall, latency, GPU memory
→ State the tradeoff you're accepting out loud

Five modules, three tiers each (Beginner foundations → Intermediate mechanisms → Architect-level system design):
🔹 RAG: hybrid search, Reciprocal Rank Fusion, re-rankers, "lost in the middle," and designing retrieval for 500M+ docs under 200ms.
🔹 Embeddings & Vector DBs: anisotropy, HNSW, product quantization, Matryoshka embeddings, sharding to a billion vectors, zero-downtime re-indexing.
🔹 Agents & Agentic Systems: the agent loop, tools, memory design, MCP, multi-agent orchestration, long-horizon autonomy, security.
🔹 Fine-Tuning: SFT vs alignment, LoRA/QLoRA/DPO, the GPU memory math, ZeRO and FSDP, serving many fine-tunes at once.
🔹 Evaluation: LLM-as-a-judge and its failure modes, faithfulness and context precision/recall, benchmark contamination, catching silent quality decay in prod.

Every answer comes with background, a worked example, and real code. The trickier claims are checked against the original papers.

It's for anyone targeting AI Engineer, ML Engineer, Applied Scientist, or the newer LLM/GenAI Engineer roles, and for practitioners who'd rather understand the machinery than copy recipes.

One tip: try to answer each question out loud before watching. The spots where you get stuck are the useful ones.

Course link here: www.udemy.com/course/genai-engineer-interview-prep…

Checkout the early bird prices valid for next 5 days

If a GenAI question has ever ended an interview for you, drop it below. I'm collecting the ones that keep coming up.

Note : This is a paid course on Udemy. This course is a collection of my Youtube Videos along with some additional content. If you prefer an ad free course like experience checkout the course

1 month ago | [YT] | 5

AI WITH Rithesh

Hey everyone! 👋

I am working to build a more engaged community of people genuinely interested in educational content. If you have been inactive or no longer watch our videos, I' would appreciate it if you could unsubscribe.

This helps me focus on connecting with viewers who are actively learning and engaging with my r channel. It keeps my metrics honest and helps us understand our real audience better.

If you have just been busy or lost interest but want to jump back in, no worries—you can always re-subscribe anytime! But if you are not engaged, hitting unsubscribe helps us both out.

Thanks for understanding! ✌️

1 month ago | [YT] | 6