My missions
→ Lead & ship production-grade AI/ML systems - including AI Agents & RAG that actually move business metrics.
→ Educate AI/ML practitioners on how real-world production AI/ML systems are designed, deployed, and maintained (the parts most people skip).

My portfolio
→ Senior Data Scientist / AI Engineer (8+ years)
→ Built an AI Agent–powered FAQ chatbot (LangChain/LangGraph, AWS) used by thousands; significantly reduced customer support load.
→ Built end-to-end RAG: ingestion, semantic/recursive chunking, retrieval strategies, GDPR-first data flows
→ Built CLV forecasting, recommendation systems, marketing optimization, and scalable MLOps systems with significant business impact

My newsletter
→ JAM with AI - read by 26,000+ AI/ML builders
→ jamwithai.substack.com

Work with me
→ Collaborations, talks, and mentoring.

I've read 25+ AI/ML books in 8 years. 

Only 11 would actually move the needle in production.

(I've shipped Data Science, MLOps, RAG systems, and AI agents)

--------
🚨 𝗕𝗲𝗳𝗼𝗿𝗲 𝘆𝗼𝘂 𝘀𝘁𝗮𝗿𝘁:
That’s why I created a complete production RAG course built from actual engineering battles, not theory, not toy examples.

If you want to learn RAG the way it works in real companies, start here:

🔗 Course link: https://lnkd.in/dzdsTWz9

Then use the following books to reinforce the fundamentals.
Learn by building them!
--------

Here's are my recommendations:
📌 𝗠𝗮𝘁𝗵𝗲𝗺𝗮𝘁𝗶𝗰𝗮𝗹 𝗙𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻𝘀
→ Mathematics of Machine Learning - Tivadar Danka
- Understand the foundations with great intuitions!
🔗 Link - https://lnkd.in/d3g494VJ

→ ML with PyTorch and Scikit-Learn - Sebastian Raschka, PhD
- Bridge theory to practice with both classical ML and deep learning.
🔗 Link -  https://lnkd.in/dkBwkjBc

📌 𝗣𝗿𝗼𝗴𝗿𝗮𝗺𝗺𝗶𝗻𝗴 & 𝗦𝘁𝗮𝘁𝘀
→ Fluent Python - Luciano Ramalho
- Write production-grade Python code.
🔗 Link - https://lnkd.in/dPzSFYmr
(skip it if you think you can learn better through courses)

→ Practical Statistics for Data Scientists - Bruce & Bruce
- A/B test models and detect data drift.
🔗 Link - https://lnkd.in/dwtNNU4R

📌 𝗡𝗟𝗣 & 𝗟𝗮𝗻𝗴𝘂𝗮𝗴𝗲
→ Python NLP Cookbook - Antić & Chakravarty
- Classical NLP for when RAG retrieves garbage.
🔗 Link -  https://lnkd.in/dVPGPew6

→ NLP with Transformers - Tunstall, von Werra, Wolf
- Understand what's under the hood of LLMs.
🔗 Link - https://lnkd.in/dqk62AQT

📌 𝗠𝗟 𝗦𝘆𝘀𝘁𝗲𝗺𝘀
→ Designing Machine Learning Systems - Chip Huyen
- Be in the 13% that reach production.
🔗 Link - https://lnkd.in/dPXxKNgk

→ ML System Design Interview - Aminian & Xu
- Think in patterns: scale from 1K to 100M users.
🔗 Link - https://lnkd.in/d_V98BW9

📌 𝗟𝗟𝗠𝘀 & 𝗔𝗴𝗲𝗻𝘁𝘀
→ AI Agents in Practice - Valentina Alto
- Build agents that work in production.
🔗 Link -https://lnkd.in/d92bycbU

→ AI Engineering - Chip Huyen
- LLMs in production: RAG, prompts, infrastructure.
🔗 Link - https://lnkd.in/dw_F-PtH

📌 𝗜𝗻𝗳𝗿𝗮𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲
→ Designing Data-Intensive Applications - Martin Kleppmann
- Your model is only as good as your infrastructure.
🔗 Link - https://lnkd.in/duWW2tD6

𝗠𝘆 𝗵𝗼𝗻𝗲𝘀𝘁 𝘁𝗮𝗸𝗲:
Pick 1-2 books. 
Build something small. 
Come back when you hit problems.

Detailed post on WHEN each saved me → Link in comments 👇

𝗪𝗵𝗮𝘁 𝗯𝗼𝗼𝗸𝘀 𝘀𝗵𝗮𝗽𝗲𝗱 𝘆𝗼𝘂𝗿 𝗷𝗼𝘂𝗿𝗻𝗲𝘆?

PS: There are many books out there, but these are the list of must read from my end. For learning LLMs as well there are multiple resources, so didnt include. 

--
♻️ Repost if you found it helpful 💚 
➕Join 31.000+ AI/ML builders here: https://lnkd.in/ds_SzEUH

Shirin Khosravi Jam

With a fondness for mathematics, my academic and professional journey led me to data science, where I've had the opportunity to work on projects that have positively impacted various Businesses. My graduate studies in Germany allowed me to delve into areas like finance and Data Science, where I could apply Machine Learning to practical challenges.

In my toolset, you'll find skills in Python, SQL, MLOps, AWS, A/B testing, and Big Data technologies. I approach my daily tasks with a focus on collaboration, aiming to ensure data accuracy and meaningful insights.

I value continuous learning and open dialogue with peers in the industry. If you're interested, I'd welcome the chance to talk and learn from each other.

Educator in Mathematics & Machine Learning | Author of 'Mathematics of Machine Learning' | Making complex concepts accessible through visual storytelling

ML/AI research engineer. Author of Build a Large Language Model From Scratch (amzn.to/4fqvn0D) and Ahead of AI (magazine.sebastianraschka.com), on how LLMs work and the latest developments in the field.

O’Reilly author (Fluent Python, 2 editions in 9 languages)

Building something new | AI x storytelling x education

AI and App Innovation Technical Architect at Microsoft | Tech Author | MSc in Data Science

Associate Professor in Cambridge. Decentralised systems & local-first software. Author of Designing Data-Intensive Applications

Sr. Data Scientist/ AI Engineer | Data Science, RAG, AI Agents, & MLOps | Germany’s Top Female Voice in AI 🇩🇪

Ilya Chukhlyaev

I'm a Data Scientist and self-taught RAG & AI agents 
Shipping in production 12+ months... 

Most roadmaps miss what actually breaks in production
I learned this the hard way...

This is from actual production experience...
let's share and help as many as people. 😇 

--------
🚨 𝗕𝗲𝗳𝗼𝗿𝗲 𝘆𝗼𝘂 𝘀𝘁𝗮𝗿𝘁:
I built a complete production grade RAG course 
learning from these production battles.
→ Real AI system architecture (not tutorials) 

🔗 Course link: https://lnkd.in/dzdsTWz9 

Start with the course and use below resources to learn further.
--------

⏩ 𝗣𝘆𝘁𝗵𝗼𝗻 𝘀𝗸𝗶𝗹𝗹𝘀
✅ 𝗙𝗮𝘀𝘁𝗔𝗣𝗜 
– Build lightweight, production-ready APIs in no time
🔗 Video1: https://lnkd.in/dpADd3gh
🔗 Video2: https://lnkd.in/dcRmk2sk

✅ 𝗔𝘀𝘆𝗻𝗰 𝗣𝗿𝗼𝗴𝗿𝗮𝗺𝗺𝗶𝗻𝗴 
– Speed up agents by handling multiple tasks asynchronously
🔗 Video: https://lnkd.in/dQDtam6S

✅ 𝗣𝘆𝗱𝗮𝗻𝘁𝗶𝗰 
– Data validation and settings management that keeps your agent robust and reliable
🔗 Video: https://lnkd.in/dArUdWKT

✅ 𝗟𝗼𝗴𝗴𝗶𝗻𝗴 
– Debugging agents without good logs? Nearly impossible.
🔗 Video: https://lnkd.in/dYP3tWAk

✅ 𝗧𝗲𝘀𝘁𝗶𝗻𝗴 
– Unit and integration tests will save you when agents become complex.
🔗 Unit test: https://lnkd.in/d-sYNxxz
🔗 Integration test: https://lnkd.in/d6waJuGp

✅ 𝗗𝗮𝘁𝗮𝗯𝗮𝘀𝗲 𝗠𝗮𝗻𝗮𝗴𝗲𝗺𝗲𝗻𝘁 (SQLAlchemy + Alembic) 
– Essential for RAG and storing agent memories
🔗 Video: https://lnkd.in/dj9UsCjd

⏩ 𝗥𝗔𝗚 𝘀𝗸𝗶𝗹𝗹𝘀
✅ 𝗨𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱𝗶𝗻𝗴 𝗥𝗔𝗚 
– Learn what RAG is and why it matters
🔗 Video - https://lnkd.in/dGUijEMw

✅ 𝗧𝗲𝘅𝘁 𝗘𝗺𝗯𝗲𝗱𝗱𝗶𝗻𝗴𝘀 
– Foundation of search and retrieval in RAG systems
🔗 Video - https://lnkd.in/dtG7m-mv

✅ 𝗩𝗲𝗰𝘁𝗼𝗿 𝗗𝗮𝘁𝗮𝗯𝗮𝘀𝗲 
– Store and retrieve embeddings efficiently
🔗 Video - https://lnkd.in/dqbrK3d7

✅ 𝗖𝗵𝘂𝗻𝗸𝗶𝗻𝗴 𝗦𝘁𝗿𝗮𝘁𝗲𝗴𝗶𝗲𝘀 
– Split data smartly for better retrieval quality
🔗 Video - https://lnkd.in/dVRPRVfN

✅ 𝗥𝗔𝗚 𝘄𝗶𝘁𝗵 𝗣𝗼𝘀𝘁𝗴𝗿𝗲𝗦𝗤𝗟
– RAG without expensive vector DBs? Yes, possible.
🔗 Video - https://lnkd.in/dDm7miwh

✅ 𝗥𝗔𝗚 𝘄𝗶𝘁𝗵 𝗟𝗮𝗻𝗴𝗖𝗵𝗮𝗶𝗻 
– Chain together retrieval, LLMs, and prompts easily
🔗 Video - https://lnkd.in/d3PuPKbF

✅ 𝗥𝗔𝗚 𝗘𝘃𝗮𝗹𝘂𝗮𝘁𝗶𝗼𝗻𝘀 
– Know if your retrieval and answers are actually good
🔗 Video - https://lnkd.in/dn-NDF_U 

✅ 𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝗱 𝗥𝗔𝗚 𝗧𝗲𝗰𝗵𝗻𝗶𝗾𝘂𝗲𝘀 
– Fine-tune your RAG system for real-world performance
🔗 Video - https://lnkd.in/dcXwX5Pp 

⏩ 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁 𝗙𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸
✅ 𝗟𝗮𝗻𝗴𝗚𝗿𝗮𝗽𝗵 
– This one is mainly used in organisations to build production ready AI Agents.
🔗 Video - https://lnkd.in/due46xmV

✅ 𝗣𝗿𝗼𝗺𝗽𝘁 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝗴𝘂𝗶𝗱𝗲 
– It's also good to follow specific guides from the LLM providers.
🔗 Link - https://lnkd.in/dAMk64iC 

✅ Langfuse (YC W23) 
– Definitely getting more popular now!
🔗 Video - https://lnkd.in/dQbpvSVv

--
♻️ Repost if you found it helpful 💚 
➕Join 31.000+ AI/ML builders here: https://lnkd.in/ds_SzEUH

Data Scientists! go beyond Notebooks!

FREE 25-day MLOps course covers everything:

𝗙𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻𝘀:
↳ Git & GitHub (version control basics)
↳ OOPs in Python (clean code structure)
↳ DVC (data versioning done right)

𝗖𝗼𝗿𝗲 𝗠𝗟𝗢𝗽𝘀 𝗧𝗼𝗼𝗹𝘀:
↳ MLflow (experiment tracking)
↳ Docker (containerization)
↳ CI/CD pipelines (automation)

𝗖𝗹𝗼𝘂𝗱 & 𝗜𝗻𝗳𝗿𝗮𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲:
↳ AWS S3 integration
↳ MongoDB setup
↳ FastAPI for ML apps

𝗣𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝗦𝘆𝘀𝘁𝗲𝗺𝘀:
↳ Kubernetes (container orchestration)
↳ Prometheus + Grafana (monitoring)
↳ EKS cluster deployment

𝟮 𝗖𝗼𝗺𝗽𝗹𝗲𝘁𝗲 𝗣𝗿𝗼𝗷𝗲𝗰𝘁𝘀:
↳ Vehicle Insurance ML Pipeline
↳ Full Capstone with end-to-end deployment

Total runtime: 𝟯𝟬+ 𝗵𝗼𝘂𝗿𝘀 of hands-on content.
Zero cost. Production-grade skills.

PS: I haven't gone through the course myself. 
But I was surprised with the list of tools used in here!
This is exactly what I mainly learnt and used in production.
It's really worth to check it out!
Great job by Vikash Das on YouTube.

Thanks for sharing it Ravi Shankar!

--
♻️ Save this. Share with someone learning MLOps.
➕ Join 33.000+ serious AI/ML builders here: https://lnkd.in/ds_SzEUH

Data Science Professional | MLOps Specialist | Providing 1-1 Mentorship

Engineering Manager, ML

Sr. Data Scientist/ AI Engineer | Data Science, RAG, AI Agents, & MLOps | Germany’s Top Female Voice in AI (Favicon) 🇩🇪

Gaurav Bhattacharya

The only resources you need to master LLMs in 2025/2026
85% covered by just two books!
Another 5% with below videos 👇

And the last 19% (most important) 
That’s the magic you unlock by actually building 😄

If you’re serious about real-world LLM applications, start here:

1️⃣ 𝗡𝗟𝗣 𝘄𝗶𝘁𝗵 𝗧𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺𝗲𝗿𝘀 by Lewis Tunstall, Leandro von Werra and Thomas Wolf
A solid foundation before diving into LLMs. Its highly important to know the pre-LLM era - comes very handy when working on Language data.
🔗 Book link - https://lnkd.in/diUjyCeg

2️⃣ 𝗛𝗮𝗻𝗱𝘀-𝗢𝗻 𝗟𝗮𝗿𝗴𝗲 𝗟𝗮𝗻𝗴𝘂𝗮𝗴𝗲 𝗠𝗼𝗱𝗲𝗹𝘀 by Jay Alammar & Maarten Grootendorst
From embeddings to RAG to fine-tuning - complete coverage
🔗 Book link - https://lnkd.in/ddb7kZ7D
🔗 Code link - https://lnkd.in/de52uRmQ

3️⃣ 𝗦𝘁𝗮𝗻𝗳𝗼𝗿𝗱 𝗖𝗦𝟮𝟱 – 𝗕𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝗟𝗮𝗿𝗴𝗲 𝗟𝗮𝗻𝗴𝘂𝗮𝗴𝗲 𝗠𝗼𝗱𝗲𝗹𝘀 (𝗟𝗟𝗠𝘀)
🔗 Link - https://lnkd.in/dpPG_VSz

4️⃣ 𝗧𝗵𝗲 𝗹𝗲𝗴𝗲𝗻𝗱𝗮𝗿𝘆 - 𝗨𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱𝗶𝗻𝗴 𝗟𝗟𝗠𝘀 𝗹𝗶𝗸𝗲 𝗖𝗵𝗮𝘁𝗚𝗣𝗧 by 𝗔𝗻𝗱𝗿𝗲𝗷 𝗞𝗮𝗿𝗽𝗮𝘁𝗵𝘆
🔗 Link - https://lnkd.in/d5v92ZCA

5️⃣ 𝗙𝗶𝗻𝗲-𝘁𝘂𝗻𝗶𝗻𝗴 𝗟𝗟𝗠𝘀 𝗼𝗻 𝗖𝘂𝘀𝘁𝗼𝗺 𝗗𝗮𝘁𝗮 - 𝗤𝘂𝗮𝗻𝘁𝗶𝘇𝗮𝘁𝗶𝗼𝗻, 𝗟𝗼𝗥𝗔, 𝗤𝗟𝗼𝗥𝗔 by Krish Naik
Part 1: 🔗 - https://lnkd.in/dsgPtSnw
Part 2: 🔗 - https://lnkd.in/d9vbREdB

This is more than enough to get hands-on and start building!
Don't get stuck into this forever!

Which ones have you explored?
Or what else would you recommend? 👇

---
🚨 If you are starting with AI engineering:
If you want to learn RAG the way it works in real-world, start here:

🔗 Course link: https://lnkd.in/dzdsTWz9
➕ Join 31.500+ AI/ML builders here: https://lnkd.in/ds_SzEUH

Post-training LLMs at 🤗

Head of Research at Hugging Face

Co-founder at 🤗 Hugging Face – Angel

LLM Builder, O'Reilly Author

Data Scientist | Psychologist | Author of "Hands-On Large Language Models"

Youtuber (1.2 Million+ Subscribers)

Raju Jadhav

70% of ML engineers fail system design interviews.

4 Videos . 1 Github Repo . 3 Books to help you 👇 

🔹Youtube videos:
1. How to Master ML System Design
→ https://lnkd.in/d89ywdkp
→ Step by step guide (went through it - its good)

2. Stanford MLSys Seminars (playlist)
→ https://lnkd.in/gP3UDqwn 
→ How Netflix/Uber actually scale

🔹Interview examples:
1. Spotify ML Question - Design a Recommendation System 
→ https://lnkd.in/dr87ADq8

2. Instagram ML Question - Design a Ranking Model 
→ https://lnkd.in/dqMxqgNc

🔹 Github Repo:
1. System Design Primer 
→ https://lnkd.in/gmAg4nBb 
→ Master the patterns that matter

🔹 Book recommendations:
1. Designing ML Systems: https://amzn.to/41V1n9t
2. AI Engineering: https://amzn.to/43Fa1u8
3. ML System Design Interview: https://amzn.to/3YGed9n

Hope it is useful!
Image credit: Chip Huyen
 
--
♻️ Repost to help other engineers 💚 
➕ Join 33.000+ serious AI/ML builders here: https://lnkd.in/ds_SzEUH

89.999+ academic papers exists
9 papers that shaped modern AI
every AI/ML engineer should read them 👇

Most developers jump straight into LLMs without understanding the foundational breakthroughs.

Here's your reading roadmap ↓

1️⃣ 𝐄𝐟𝐟𝐢𝐜𝐢𝐞𝐧𝐭 𝐄𝐬𝐭𝐢𝐦𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐖𝐨𝐫𝐝 𝐑𝐞𝐩𝐫𝐞𝐬𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧𝐬 𝐢𝐧 𝐕𝐞𝐜𝐭𝐨𝐫 𝐒𝐩𝐚𝐜𝐞 (𝟐𝟎𝟏𝟑)
Where it all began.
Introduced word2vec and semantic word understanding.
→ Made "king - man + woman = queen" math possible
→ 70K+ citations, still used everywhere today
🔗 https://lnkd.in/dF4KBQWW

2️⃣ 𝐀𝐭𝐭𝐞𝐧𝐭𝐢𝐨𝐧 𝐈𝐬 𝐀𝐥𝐥 𝐘𝐨𝐮 𝐍𝐞𝐞𝐝 (𝟐𝟎𝟏𝟕)
Killed RNNs. Created the Transformer architecture.
→ Every major LLM uses this foundation
🔗 https://lnkd.in/d6trTxgs

3️⃣ 𝐁𝐄𝐑𝐓 (𝟐𝟎𝟏𝟖)
Stepping stone on Transformer architecture. Introduced bidirectional pretraining for deep language understanding.
→ Looks left AND right to understand meaning
🔗 https://lnkd.in/dv8YE43j

4️⃣ 𝐆𝐏𝐓 (𝟐𝟎𝟏𝟖)
Unsupervised pretraining + supervised fine-tuning.
→ Started the entire GPT revolution
🔗 https://lnkd.in/dkadsJXk

5️⃣ 𝐂𝐡𝐚𝐢𝐧-𝐨𝐟-𝐓𝐡𝐨𝐮𝐠𝐡𝐭 𝐏𝐫𝐨𝐦𝐩𝐭𝐢𝐧𝐠 (𝟐𝟎𝟐𝟐)
"Think step by step" = 3x better reasoning
🔗 https://lnkd.in/dCNJwTrD

6️⃣ 𝐒𝐜𝐚𝐥𝐢𝐧𝐠 𝐋𝐚𝐰𝐬 𝐟𝐨𝐫 𝐍𝐞𝐮𝐫𝐚𝐥 𝐋𝐚𝐧𝐠𝐮𝐚𝐠𝐞 𝐌𝐨𝐝𝐞𝐥𝐬 (𝟐𝟎𝟐𝟎)
Math behind "bigger = better"
→ Predictable power laws guide AI investment
🔗 https://lnkd.in/dfnniFVB

7️⃣ 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐭𝐨 𝐒𝐮𝐦𝐦𝐚𝐫𝐢𝐳𝐞 𝐰𝐢𝐭𝐡 𝐇𝐮𝐦𝐚𝐧 𝐅𝐞𝐞𝐝𝐛𝐚𝐜𝐤 (𝟐𝟎𝟐𝟎)
Introduced RLHF - the secret behind ChatGPT's helpfulness
🔗 https://lnkd.in/dwkWVrUP

8️⃣ 𝐋𝐨𝐑𝐀 (𝟐𝟎𝟐𝟏)
Fine-tune 175B models by training 0.01% of weights
→ Made LLM customization affordable for everyone
🔗 https://lnkd.in/dQ4KKwXU

9️⃣ 𝐑𝐞𝐭𝐫𝐢𝐞𝐯𝐚𝐥-𝐀𝐮𝐠𝐦𝐞𝐧𝐭𝐞𝐝 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧 (𝟐𝟎𝟐𝟎)
Original RAG paper - combines retrieval with generation
→ Foundation of every knowledge-grounded AI system
🔗 https://lnkd.in/dWhkp3jG

Which of these papers have you read?
Any other paper you want to add to this list?

♻️ Repost to share with your network 💚
➕ Join 31.000+ AI/ML builders here: https://lnkd.in/ds_SzEUH

12 repos ML/AI Engineer should definitely know
(Covering ML, RAG, PyTorch, MLOps, Agents)
Bookmark them! 👇 

1️⃣ 𝗠𝗮𝗰𝗵𝗶𝗻𝗲 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗳𝗼𝗿 𝗕𝗲𝗴𝗶𝗻𝗻𝗲𝗿𝘀 𝗯𝘆 𝗠𝗶𝗰𝗿𝗼𝘀𝗼𝗳𝘁
→ Start here if you’re brand new to ML.
Covers basic ML concepts in Jupyter notebooks beginner-friendly and visual.
🔗 https://lnkd.in/gGithBVP

2️⃣ 𝗟𝗲𝗮𝗿𝗻 𝗣𝘆𝗧𝗼𝗿𝗰𝗵 𝗳𝗼𝗿 𝗗𝗲𝗲𝗽 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴
→ A great repo to learn PyTorch - could be a bit outdated but the concepts still applies.
🔗 https://lnkd.in/d4ptSuUY

3️⃣ 𝗛𝗮𝗻𝗱𝘀-𝗼𝗻 𝗟𝗮𝗿𝗴𝗲 𝗟𝗮𝗻𝗴𝘂𝗮𝗴𝗲 𝗠𝗼𝗱𝗲𝗹𝘀
→ This repo supports the Hands-On LLM book.
Learn everything from basic language models to finetuning with real notebooks.
🔗 https://lnkd.in/gpvPemeG

4️⃣ 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁𝘀 𝗳𝗼𝗿 𝗕𝗲𝗴𝗶𝗻𝗻𝗲𝗿𝘀
→ A fantastic beginner-friendly course to get started with AI agents.
Free 11-lesson hands-on curriculum - no fluff, just code.
🔗 https://lnkd.in/gVm5gmv7

5️⃣ 𝗣𝗿𝗼𝗺𝗽𝘁 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝗚𝘂𝗶𝗱𝗲
→ One-stop-shop for prompt engineering.
Guides, papers, lectures, and tons of curated examples.
🔗 https://lnkd.in/gDt3Sknr

6️⃣ 𝗟𝗟𝗠 𝗖𝗼𝘂𝗿𝘀𝗲
→ Hands-on course covering the entire LLM lifecycle — design to deployment.
Includes roadmaps + Colab notebooks.
🔗 https://lnkd.in/gUPJmWNM

6️⃣ 𝗚𝗲𝗻𝗔𝗜 𝗔𝗴𝗲𝗻𝘁𝘀
→ Great tutorials + code for building agent-based LLM systems.
Covers everything from simple tool-using agents to advanced workflows.
🔗 https://lnkd.in/gaiZgzpA

7️⃣ 𝗥𝗔𝗚 𝗧𝗲𝗰𝗵𝗻𝗶𝗾𝘂𝗲𝘀
→ One of the most comprehensive and dynamic collections of Retrieval-Augmented Generation (RAG) tutorials available
🔗 https://lnkd.in/dcGtjFgY

8️⃣ 𝗔𝗴𝗲𝗻𝘁𝘀 𝘁𝗼𝘄𝗮𝗿𝗱𝘀 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻
→ It is your hands-on guide to every building block of a GenAI agent stack.
🔗 https://lnkd.in/dnnkrJbE

9️⃣ 𝗔𝘄𝗲𝘀𝗼𝗺𝗲 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗔𝗜
→ Massive curated list of GenAI tools, papers, interviews, repos, and more. Use it as your personal R&D knowledge base.
🔗 https://lnkd.in/gsTk5nXx

🔟 𝗠𝗮𝗱𝗲 𝗪𝗶𝘁𝗵 𝗠𝗟
→ Covers full ML product lifecycle: from design to CI/CD and monitoring. If you’re serious about building production-grade ML systems, this is gold.
🔗 https://lnkd.in/gn2RhT_m

1️⃣1️⃣ 𝗗𝗲𝘀𝗶𝗴𝗻𝗶𝗻𝗴 𝗠𝗮𝗰𝗵𝗶𝗻𝗲 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗦𝘆𝘀𝘁𝗲𝗺𝘀
→ Summaries + code + diagrams from the popular O’Reilly book.
A must-read if you want to architect real-world ML pipelines.
🔗 https://lnkd.in/g26KNXfb

1️⃣2️⃣ 𝗝𝗮𝗺 𝘄𝗶𝘁𝗵 𝗔𝗜 (𝗕𝗼𝗻𝘂𝘀)
→ Your goto repo for hands-on guides
A must check out repo for production RAG Systems, AI Agents, MLOps and much more
🔗 https://lnkd.in/dr4g3jcC

What else would you add here?
Comment below 👇 

--
♻️ Repost to help other engineers 💚 
➕ Join 32.500+ serious AI/ML builders here: https://lnkd.in/ds_SzEUH

Learning ML Systems is a long term investment.
This ONE github repo + book by a Harvard Prof.
teaches you 6 pillars of ML Systems 👇 

𝗧𝗵𝗲 𝗿𝗲𝘀𝗼𝘂𝗿𝗰𝗲:
Machine Learning Systems by Prof. Vijay Janapa Reddi
Harvard's CS249r - completely free, open source.

𝗧𝗼𝗽𝗶𝗰𝘀 𝗰𝗼𝘃𝗲𝗿𝗲𝗱:
→ Scalable system architecture
→ Data pipeline design
→ Production deployment
→ Real world MLOps
→ Resource-constrained ML
→ Privacy and sustainability

No fluff. No hype.
Just the 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝗳𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻𝘀 that matter.

𝗟𝗶𝗻𝗸:
1. Book - https://lnkd.in/dacfPQqY
2. Github - https://lnkd.in/dkDHPmWm

I really hope this is helpful 🫶 

--
♻️ Repost to help other engineers 💚 
➕ Join 32.500+ serious AI/ML builders here: https://lnkd.in/ds_SzEUH

Professor at Harvard University

Daniel Ifediba

6 Interview Question and Answers of FastAPI

(you should know them now)

𝗤𝟭: 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗙𝗮𝘀𝘁𝗔𝗣𝗜 𝗯𝘂𝗶𝗹𝘁 𝗼𝗻?
FastAPI is built on two key components:
→ Starlette: Handles web requests and async event loops
→ Pydantic: Validates and parses your data

This combination gives you:
• Super fast I/O operations
• Strict type checking automatically
• Built-in data validation

The interviewer wants to know you understand it's not just a framework.
It's a well-designed combination of proven tools.

𝗤𝟮: 𝗪𝗵𝗮𝘁'𝘀 𝘁𝗵𝗲 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝗰𝗲 𝗯𝗲𝘁𝘄𝗲𝗲𝗻 𝗮𝘀𝘆𝗻𝗰 𝗱𝗲𝗳 𝘃𝘀 𝗱𝗲𝗳?
Async is the core principle for performance in FastAPI.

Use async def when your endpoint does:
• Database calls
• External API requests
• File read/write operations
• Any I/O-bound task

Regular def blocks the entire app while waiting.
Async def lets the server handle other requests while waiting.
That's where you get real performance gains.

𝗤𝟯: 𝗛𝗼𝘄 𝗱𝗼 𝘆𝗼𝘂 𝗵𝗮𝗻𝗱𝗹𝗲 𝗗𝗕 𝗰𝗼𝗻𝗻𝗲𝗰𝘁𝗶𝗼𝗻𝘀?
Dependency injection is the secret sauce in FastAPI.

With Depends, every request automatically gets:
• A fresh database session
• Proper connection management
• Clean separation of concerns

You can use the same pattern for:
• Authentication checks
• Configuration loading
• Cache client injection

The benefit: Your endpoints stay clean, testable, and decoupled.
Show you understand the pattern, not just the syntax.

𝗤𝟰: 𝗛𝗼𝘄 𝗱𝗼𝗲𝘀 𝗙𝗮𝘀𝘁𝗔𝗣𝗜 𝘃𝗮𝗹𝗶𝗱𝗮𝘁𝗲 𝗶𝗻𝗽𝘂𝘁𝘀?
Pydantic handles validation automatically.

You define a BaseModel with your expected fields.
FastAPI then:
• Auto-validates incoming requests
• Auto-serializes responses
• Returns clear error messages if validation fails

This means:
→ No more manual checks like "if 'name' not in payload"
→ Safer APIs with guaranteed data types
→ Cleaner, more maintainable code

The interviewer wants to see you understand automatic validation is a production advantage.

𝗤𝟱: 𝗪𝗵𝗮𝘁 𝗮𝗯𝗼𝘂𝘁 𝗔𝗣𝗜 𝗱𝗼𝗰𝘂𝗺𝗲𝗻𝘁𝗮𝘁𝗶𝗼𝗻?
FastAPI auto-generates interactive docs at /docs.

It uses your Python type hints to create:
• Swagger UI automatically
• Always in-sync documentation
• Interactive testing interface

Why this matters:
→ Frontend teams can test endpoints without asking you
→ New developers understand APIs immediately
→ Your documentation never goes stale

Mention: "This saves hours of manual documentation work and keeps contracts in sync."

𝗤𝟲: 𝗛𝗼𝘄 𝗱𝗼 𝘆𝗼𝘂 𝗵𝗮𝗻𝗱𝗹𝗲 𝘀𝗹𝗼𝘄 𝗼𝗽𝗲𝗿𝗮𝘁𝗶𝗼𝗻𝘀?
Background tasks keep your API responsive.

Use BackgroundTasks to offload:
• Email sending
• Log processing
• Model retraining triggers
• Heavy computations

The flow:
1. Return response to user immediately
2. Process slow tasks in background
3. User doesn't wait

This keeps your app snappy and user experience smooth.

♻️ Repost to help other engineers 💚 
➕ Join 32.500+ serious AI/ML builders here: https://lnkd.in/ds_SzEUH

kuntal patel

Everyone's running towards AI Engineering.

Meanwhile, I feel Data Science roles are a big miss.

All the traditional problems still need solving:
→ Regression
→ Classification  
→ Time Series
→ Demand Forecasting, etc
Across different domains - Marketing, Customer Analytics, etc.

Data Science is not prompt engineering or calling a LLM API.

It's about:
→ Understanding the domain and business problem deeply
→ Cleaning, wrangling, interpreting raw data
→ Asking the right questions before modeling
→ Building groundtruth & robust statistical foundations
→ Designing experiments, not just tuning parameters
→ Translating insights into business impact
→ Making the models explainable to support business, etc.

I understand that now Data Scientists need to move beyond notebooks. 
You will definitely not regret learning and building e2e ML projects.

Working on both side, I can definitely say:
There is a lot of work to be done on both sides!

Both paths lead to great careers.
Build depth, not just breadth.

Pick the domain that excites you.
Don't just go with the Hype.

True Data Science is here to stay!
Do you agree?

--
♻️ Repost if you agree 💚 
➕ Join 33.500+ serious AI/ML builders here: https://lnkd.in/ds_SzEUH

taplio

Shirin Khosravi Jam's Linkedin Analytics