updated docs

deepaksood619 · Dec 7, 2023 · c046b7e · c046b7e
1 parent 9698133
commit c046b7e
Show file tree

Hide file tree

Showing 12 changed files with 111 additions and 5 deletions.
diff --git a/docs/about-me/paper-publications-references.md b/docs/about-me/paper-publications-references.md
@@ -14,6 +14,8 @@ https://www.facebook.com/IIITDelhi/posts/2279503818733302 (Project was awarded 1
 
 ## References
 
+### College
+
 - Dr. Pushpendra Singh, Associate Dean (IRD)
 - Indraprastha Institute of Information Technology (IIIT) Delhi
 - Mobile Number - +91 88264 58886
@@ -26,3 +28,28 @@ https://www.facebook.com/IIITDelhi/posts/2279503818733302 (Project was awarded 1
 - EA: Ms. Nayana Samuel ([email protected])
 - Mobile Number - +91 96707 73311
 - Email - [email protected]
+
+### Professional
+
+#### Someone you have reported to
+
+1. Parikshit Chitalkar - CTO - Stashfin - [[email protected]](mailto:[email protected])
+ 1. My Manager at Stashfin - Sep 2020 - Feb 2023
+2. Jawahar Sukumar - Product and Data Analytics Head - ODF, Ex - Stashfin, PayU  - [[email protected]](mailto:[email protected])
+ 1. My Manager at Stashfin when I was Senior Data Engineer - Feb 2020 - Aug 2020
+ 2. Have worked with him later on after that, on a few more projects
+
+#### A peer you have collaborated with very closely
+
+1. Vishal Goyal - Senior Software Developer - Paypay Japan, Ex - Stashfin - Sep 2020 - Feb 2023
+ 1. Was working directly with him closely in all projects at Stashfin
+2. Jawahar Sukumar - Product and Data Analytics Head - ODF, Ex - PayU - [[email protected]](mailto:[email protected])
+ 1. My Manager at Stashfin when I was Senior Data Engineer - Feb 2020 - Aug 2020
+ 2. Have worked with him later on after that, on a few more projects
+
+#### Someone you have managed directly
+
+1. Nishant Verma - Cred, Ex - Stashfin
+ 1. Hired and managed directly in Stashfin
+2. Ayush Yadav - Nykaa, Ex - Stashfin
+ 1. Hired and managed directly in Stashfin
diff --git a/docs/about-me/social-links.md b/docs/about-me/social-links.md
@@ -31,6 +31,7 @@
 - Twitter - https://twitter.com/deepaksood619
 - Skype Username - [email protected]
 - Github Personal - https://github.com/deepaksood619
+- Gitlab Personal - [Deepak Sood · GitLab](https://gitlab.com/deepaksood619)
 - Github College - https://github.com/deepak15013
 - Codepen - https://codepen.io/deepaksood619
 - Project collection list - https://codepen.io/collection/AWRYzb

diff --git a/docs/ai/data-science/questions.md b/docs/ai/data-science/questions.md
@@ -129,16 +129,39 @@ The F1-score balances both precision and recall, so it's a good measure of class
 
 ## SQL
 
-- Write a SQL query to **Find the name of the student getting second highest marks in his class**
+#### Write a SQL query to Find the name of the student getting second highest marks in his class
 
+```sql
+-- with max(marks), cannot use name
 SELECT name, MAX(marks) AS max_marks
 FROM student
 WHERE marks < (SELECT MAX(marks)
 FROM student) LIMIT 1;
+-- Error - You tried to execute a query that does not include the specified expression 'name' as part of an aggregate function.
 
-SELECT marksFROM (SELECT marksFROM student ORDER BYmarksDESC LIMIT 2) AS Std ORDER BYmarksLIMIT 1;
+-- names not selected
+SELECT marks FROM (SELECT marks FROM student ORDER BY marks DESC LIMIT 2) AS Std ORDER BY marks LIMIT 1;
 
-select max(marks) from student where marks < select max(marks) from student;
+SELECT max(marks) FROM student WHERE marks < SELECT max(marks) FROM student;
+
+-- working
+select name,sal from emp where sal = (select max(sal) from emp where sal < (select max(sal) from emp));
+
+-- using window function
+with cte (
+select ROW_NUMBER() over(order by sal desc)rnum ,name,sal from emp )
+select * from cte where rnum = 2
+
+select * from
+  ( select sal
+  , rank() over (order by sal desc) as rnk
+  from
+   ( select distinct sal from emp )
+  )
+  where rnk = 2
+```
+
+#### Other Questions
 
 - Difference between procedures and functions in mysql
 - What are different types of indexes in relational database

diff --git a/docs/ai/libraries/ml-model-deployment.md b/docs/ai/libraries/ml-model-deployment.md
@@ -16,7 +16,9 @@ The model is deployed to an online prediction service cluster (generally contain
 
 ## MLOps
 
-## ML Lifecycle
+MLOps is an engineering discipline that aims to unify ML systems development (dev) and ML systems deployment (ops) in order to standardize and streamline the continuous delivery of high-performing models in production.
+
+### ML Lifecycle
 
 - Framing ML problems from business goals
 - Access, prepare and process data for the model
@@ -27,3 +29,5 @@ The model is deployed to an online prediction service cluster (generally contain
 https://www.freecodecamp.org/news/what-is-mlops-machine-learning-operations-explained
 
 [MLOps guide](https://huyenchip.com/mlops/)
+
+[MLOps Course – Build Machine Learning Production Grade Projects - YouTube](https://www.youtube.com/watch?v=-dJPoLm_gtE)
diff --git a/docs/ai/libraries/tools.md b/docs/ai/libraries/tools.md
@@ -62,6 +62,8 @@ https://github.com/streamlit/streamlit
 
 https://www.freecodecamp.org/news/build-12-data-science-apps-with-python-and-streamlit
 
+[Generative AI and Streamlit: A perfect match](https://blog.streamlit.io/generative-ai-and-streamlit-a-perfect-match/)
+
 ## Metaflow
 
 Metaflow is a human-friendly Python library that helps scientists and engineers build and manage real-life data science projects. Metaflow was originally developed at Netflix to boost productivity of data scientists who work on a wide variety of projects from classical statistics to state-of-the-art deep learning

diff --git a/docs/ai/llm/llm-building.md b/docs/ai/llm/llm-building.md
@@ -23,6 +23,45 @@ RAG is an AI framework for retrieving facts from an external knowledge base to g
 
 [What is retrieval-augmented generation? | IBM Research Blog](https://research.ibm.com/blog/retrieval-augmented-generation-RAG)
 
+## Others
+
+![LLM Working](../../media/llm-working.jpg)
+
+### How to train your ChatGPT
+
+#### Stage 1: Pretraining
+
+1. Download ~10TB of text
+2. Get a cluster of ~6,000 GPUs
+3. Compress the text into a neural network, pay ~$2M, wait ~12 days
+4. Obtain base model
+
+#### Stage 2: Finetuning
+
+1. Write labeling instructions
+2. Hire people (or use scale.ai!), collect 100K high quality ideal Q&A responses, and/or comparisons
+3. Finetune base model on this data, wait ~1 day
+4. Obtain assistant model
+5. Run a lot of evaluations
+6. Deploy
+7. Monitor, collect misbehaviors, go to step 1
+
+### LLM Security
+
+- Jailbreaking
+- Prompt injection
+- Backdoors & data poisoning
+- Adversarial inputs
+- Insecure output handling
+- Data extraction & privacy
+- Data reconstruction
+- Denial of service
+- Escalation
+- Watermarking & evasion
+- Model theft
+
+[[1hr Talk] Intro to Large Language Models - YouTube](https://www.youtube.com/watch?v=zjkBMFhNj_g)
+
 ## Dev Tools
 
 - [Welcome to LangChain - 🦜🔗 LangChain 0.0.180](https://python.langchain.com/en/latest/index.html)

diff --git a/docs/ai/llm/readme.md b/docs/ai/llm/readme.md
@@ -4,6 +4,8 @@
 - [Design Patterns](ai/llm/design-patterns.md)
 - [ChatGPT Prompt Engineering](ai/courses/chatgpt-prompt-eng.md)
 
+MMLU - Massive Multitask Language Understanding
+
 ## GPT-3
 
 Generative Pre-trained Transformer 3 (GPT-3; stylized GPT·3) is an autoregressive language model that uses deep learning to produce human-like text. Given an initial text as prompt, it will produce text that continues the prompt.
@@ -32,6 +34,7 @@ Moving from information to knowledge age
 
 #### AI Generators
 
+- [**Gemini** - Google DeepMind](https://deepmind.google/technologies/gemini)
 - [Stable Diffusion Online](https://stablediffusionweb.com/) - [Stability AI](https://stability.ai/) - Generation of unique, realistic, high-quality images, art, logos, and designs
 - Amazon Bedrock
 - Amazon Titan - Amazon Titan FMs are a family of models built by Amazon that are pretrained on large datasets, which makes them powerful, general-purpose models

diff --git a/docs/cloud/others/others-saas.md b/docs/cloud/others/others-saas.md
@@ -56,6 +56,7 @@ Components
 ## Test Taking / Online assessment
 
 - ThinkExam
+- **Speedexam** - https://www.speedexam.net - 50 exams per month free
 - **Coding**
   - Codility (good but high price)
   - Evalground (150 per test, 100 min no of tests, 15000)
@@ -70,7 +71,6 @@ Components
 - TCExam - https://tcexam.org/docs/installation - https://github.com/tecnickcom/tcexam
 - Papershala - https://papershala.com
 - Edbase - http://www.edbase.net
-- **Speedexam** - https://www.speedexam.net - 50 exams per month free
 - https://belong.co
 
 ## Supply Chain Management

diff --git a/docs/databases/nosql-databases/snowflake.md b/docs/databases/nosql-databases/snowflake.md
@@ -185,6 +185,10 @@ DROP WAREHOUSE IF EXISTS sf_tuts_wh;
 
 ![snowflake-sql-lifecycle-diagram](../../media/Pasted%20image%2020231205120527.png)
 
+## Others
+
+- Streamlit
+
 ## Links
 
 [The Snowflake Data Cloud - Mobilize Data, Apps, and AI](https://www.snowflake.com/en/)

diff --git a/docs/economics/mental-models/macroeconomics.md b/docs/economics/mental-models/macroeconomics.md
@@ -165,3 +165,5 @@ The **impossible trinity** (also known as **the impossible [trilemma](https://en
 - [How The UN Destroyed Iraq For Oil | Epic Economics - YouTube](https://www.youtube.com/watch?v=PM3ZMWtbBag)
 - [Watch Out! They’re LYING To You About Inflation & The Economy!! - YouTube](https://www.youtube.com/watch?v=SUE1mNNjcZc)
 - [The Thinking Behind Why Cash Is Now Good (and not Trash)](https://www.linkedin.com/pulse/thinking-behind-why-cash-now-good-trash-ray-dalio)
+- [**Post Labor Economics**: How will the economy work after AGI? Recent thoughts and conversations - YouTube](https://www.youtube.com/watch?v=eD5GlCIS0sA) - Status game, Status economy
+-
diff --git a/docs/knowledge/links.md b/docs/knowledge/links.md
@@ -1173,5 +1173,6 @@
 1163. [This Country Doesn't Care About Being Rich, They Are Measuring Happiness - YouTube](https://www.youtube.com/watch?v=a88l92ntAOM)
 1164. [The Bhopal Gas Leak | Who were Railway Men? | Dhruv Rathee - YouTube](https://www.youtube.com/watch?v=DOApK2Yf5j4)
 1165. [Citigroup and the 'financial supermarket' experiment | FT Film - YouTube](https://www.youtube.com/watch?v=3njfN0IOz38)
+1166. [You probably won’t survive 2024... Top 10 Tech Trends - YouTube](https://www.youtube.com/watch?v=vyQv563Y-fk)
 
 ## End
diff --git a/docs/media/llm-working.jpg b/docs/media/llm-working.jpg