Full-stack engineer building AI, real-time systems & scalable web apps ๐ŸŒ. Turning ideas๐Ÿ’ก โ†’ products ๐Ÿš€

Joined February 2024
134 Photos and videos
Day 64/100 of #100DaysOfCode ๐Ÿณ๐Ÿš€ Today I explored Docker Networks & Volumes. ๐Ÿ”— Networks let containers communicate ๐Ÿ’พ Volumes let containers remember data Realization: Containers alone are temporary. Networks Volumes make apps scalable & production-ready ๐Ÿ”ฅ #DevOps
2
80
Day 63/100 of #100DaysOfCode ๐Ÿš€ Today I finally understood Dockerfiles ๐Ÿณ A Dockerfile is basically a recipe for creating the SAME app environment anywhere ๐ŸŒ No more: โŒ โ€œWorks on my machineโ€ โŒ Dependency chaos โŒ Version mismatch DevOps journey begins ๐Ÿ”ฅ #Docker #NodeJS
4
124
Day 62/100 of #100DaysOfCode ๐Ÿš€ Today I explored Agentic AI ๐Ÿค– AI that doesnโ€™t just respondโ€ฆ It plans, reasons, uses tools, adapts, remembers context, and gets tasks DONE. Weโ€™re moving from: โ€œAI that answersโ€ โžก๏ธ โ€œAI that actsโ€ ๐Ÿ”ฅ #AI #AgenticAI #LLM #Coding
2
1
120
Day 61/100 of #100DaysOfCode ๐Ÿณ๐Ÿ”ฅ Today I finally understood Docker Images vs Containers. ๐Ÿ“˜ Image = Recipe/Blueprint ๐Ÿ” Container = Running meal/app One image can create multiple containers ๐Ÿš€ โ€œBuild once. Run anywhere.โ€ finally makes sense ๐Ÿ’ก #Docker #DevOps #Coding
3
106
Day 60/100 of #100DaysOfCode ๐Ÿš€Today I explored TypeScript utility types Zod inference ๐Ÿ”ฅ โœ… Pick โœ… Partial โœ… Readonly โœ… Record โœ… Exclude โœ… Mapped Types โœ… z.infer() Big realization: Great types donโ€™t just prevent bugsโ€ฆ they design better systems โšก #TypeScript #Zod
1
40
Day 59 of #100DaysOfCode: Text-to-SQL Generator ๐Ÿ’พ Logic: Dynamic Schema Extraction โžก๏ธ LLM Translation โžก๏ธ SQLite Execution. Tech: Python, LangChain, Ollama (DeepSeek-R1 local), Streamlit. Deterministic SQL with temp=0 for accuracy. ๐ŸŽฏ #AI #Python #SQL #OpenSource #GenAI
1
2
141
๐Ÿš€ Day 58 of #100DaysOfCode Today I explored AWS โ˜๏ธ Learned about EC2, S3, IAM, Route53, Lambda, RDS, DynamoDB, ECS, ECR, Bedrock, SageMaker & more. Finally understanding how scalable apps are built, deployed, secured, containerized & powered by AI ๐Ÿ”ฅ #AWS #CloudComputing
3
97
Day 57 of #100DaysOfCode ๐Ÿš€ Today I explored the real power of TypeScript: โ€ข Type Inference vs Annotations โ€ข Interfaces vs Types โ€ข Generics <T> โ€ข Union & Intersection Types Good types donโ€™t just prevent bugs โ€” they design better systems โšก #TypeScript
3
56
Day 56/100 ๐Ÿš€ LLMs donโ€™t execute toolsโ€ฆ they suggest them ๐Ÿคฏ Built tool calling in LangChain: ๐Ÿงฉ Bind tools ๐Ÿค– LLM suggests โš™๏ธ Backend executes From โ€œjust textโ€ โ†’ real-world actions ๐Ÿ”ฅ This is where AI gets serious. #100DaysOfCode #AI #LangChain #LLM #BuildInPublic
1
5
114
Day 55/100 ๐Ÿš€ Diving into TypeScript basics & finally getting type safety ๐Ÿ”Arrays โœ… Tuples โœ… Enums โœ… Big one: โš ๏ธ any = chaos ๐Ÿ›ก๏ธ unknown = safety ๐Ÿ”ฅ Always prefer unknown over any ๐Ÿ”ฅ JS โ†’ TS = guesswork โ†’ engineering ๐Ÿ’ป #100DaysOfCode #TypeScript #DevJourney
1
46
Day 54/100 ๐Ÿš€ Built an AI Image Generation App ๐ŸŒ Not just prompts โ†’ but a full system: ๐Ÿง  Smart prompt engineering โšก FastAPI backend ๐ŸŽจ Streamlit UI From text โ†’ ads โ†’ try-ons โ†’ scenes AI isnโ€™t just tools anymoreโ€ฆ Itโ€™s experiences. ๐Ÿ”ฅ #100DaysOfCode #AI #BuildInPublic
1
4
145
Day 53/100 ๐Ÿš€ Access Token = ๐ŸŽŸ๏ธ short-lived entry Refresh Token = ๐Ÿ” renews access No localStorage โŒ (XSS risk) Secure cookies memory โœ… Mind blown ๐Ÿคฏ: OTP isnโ€™t storedโ€”only its hash. Server verifies what it never โ€œknows.โ€ Security = layers, not features ๐Ÿ” #100DaysOfCode
1
2
71
Day 52/100 ๐Ÿš€ Today I explored Tools in LangChain ๐Ÿ”ง โ†’ Built-in tools (plug & play) โ†’ Custom tools (your logic) โ†’ Creating tools (function description = power) โ†’ Toolkits (bundled capabilities) LLMs are no longer just chatbotsโ€ฆ they ACT โšก #100DaysOfCode #AI #LangChain
2
95
Day 51/100 ๐Ÿš€ Learned the backbone of web security: ๐Ÿ” Hashing Salting (no plain passwords) ๐Ÿ›ก๏ธ bcrypt (slow = secure) ๐Ÿช Cookies flags (HttpOnly, Secure, SameSite) ๐Ÿ”‘ Tokens for scalable auth Security isnโ€™t optional. Itโ€™s the foundation. #100DaysOfCode #WebSecurity #WebDev
2
67
Built a YouTube chatbot with LangChain ๐Ÿคฏ ๐ŸŽฅ โ†’ Transcript โ†’ Chunks โ†’ Embeddings โ†’ FAISS โ†’ Retrieval โ†’ Prompt โ†’ LLM โ†’ Answer It doesnโ€™t โ€œwatchโ€โ€ฆ it understands ๐Ÿ” RAG in action = smarter, grounded AI ๐Ÿš€ #AI #LangChain #RAG #100DaysOfCode
1
4
160
๐Ÿš€ Day 49 of #100DaysOfCode Learned the difference between: ๐Ÿ” Authentication (Who you are) ๐Ÿ›‚ Authorization (What you can access) โœ… Validation (Is input correct?) ๐Ÿ” Verification (Is it really you?) Security = Layered Trust โšก #WebDev #Backend
3
66
Day 48/100 ๐Ÿš€ Finally understood RAG ๐Ÿคฏ Itโ€™s not about making LLMs smarterโ€ฆ Itโ€™s about giving them the right context at the right time Index โ†’ Retrieve โ†’ Augment โ†’ Generate Less hallucination โœ… More accuracy โœ… Building real AI now ๐Ÿ’ช #100DaysOfCode #AI #RAG #GenAI
2
60
๐Ÿš€ Day 47 #100DaysOfCode Revised Prisma ORM, SQL Joins & Foreign Keys ๐Ÿ’ก INNER, LEFT, RIGHT, FULL JOIN finally make sense ๐Ÿ”ฅ Big realization: Databases arenโ€™t just storageโ€ฆ theyโ€™re about relationships & connections ๐Ÿคฏ Still learning. Still building ๐Ÿš€ #SQL #Prisma #WebDev
3
107
๐Ÿš€ Day 46/100: Finally understanding how AI really works ๐Ÿคฏ RAG = AI looks up info before answering ๐Ÿ“– Fine-tuning = making AI a specialist ๐ŸŽฏ Embeddings = words โ†’ numbers ๐Ÿ”ข Big lesson: AI is only as smart as the data we give it. #100DaysOfCode #AI #MachineLearning #GenAI
3
55
Day 45/100 ๐Ÿš€ Backend concepts finally clicked ๐Ÿ”ฅ โœ” Mongoose = structure for MongoDB โœ” Relationships (1-1, 1-M, M-M) โœ” app.use(express.json()) = req.body works โœ” express.static() = serve files easily Small concepts. Massive impact. #100DaysOfCode #NodeJS #MongoDB #Backend
4
69