AIA

👨‍💻 AI & Software & DevOps Engineer | Sharing fixes & tools for smarter, faster systems | Simplicity over clever code #AI #DevOps #Learner #Fastone #TechHumor

[AIA @000ADIL]

🧵 Our Angular app had intermittent infinite loading — blank screen, no API calls, just a spinner forever.

Root cause? ChunkLoadError.

Nginx was caching index.html for 30 days. After every deploy, browsers served stale index.html pointing to old chunk files that no longer existed on the server.

Fix: Cache-Control: no-store on index.html. Cache static assets separately with max-age.

One config change. Problem gone.

Never cache index.html in a SPA. Cache the assets, not the entry point.

#Angular #DevOps #Nginx #ChunkLoadError #Frontend #Kubernetes #WebDev

[AIA @000ADIL]

@ndtvindia @grok why

[ʀᴜᴅʜʀᴀ ʏᴀᴅᴀᴠ🇮🇳 @Rudhrayadav001]

गोरखपुर से हैदराबाद जाने के लिए 5 लोग काम करने और रोज़ी-रोटी कमाने के लिए निकले थे।

उन्होंने स्लीपर बस की टिकट बुक कराने के लिए एक ट्रैवल एजेंट से संपर्क किया,

एजेंट ने प्रति व्यक्ति ₹2300 के हिसाब से कुल ₹11,500 रुपये वसूल लिए,

पैसे लेने के बाद यात्रियों को बस में बैठाने के नाम पर कुछ दूर तक ले जाया गया और फिर कुशीनगर के हरैया चौराहे के पास हाईवे पर उतार दिया गया,

उन्हें भरोसा दिलाया गया था कि स्लीपर बस जल्द ही वहीं पहुंचेगी, लेकिन काफी देर इंतजार करने के बाद भी कोई बस मौके पर नहीं आई,

जब यात्रियों ने एजेंट से संपर्क करने की कोशिश की तो उसका मोबाइल फोन बंद मिला और वह फरार हो चुका था,

इस घटना के बाद साफ हो गया कि उनके साथ धोखाधड़ी हुई है और उनकी मेहनत की कमाई ठग ली गई,

प्रशासन ऐसे फर्जी ट्रैवल एजेंटों पर सख्त कार्रवाई की जाए ताकि आम लोगों के साथ इस तरह की धोखाधड़ी न हो सके।

[AIA @000ADIL]

@shaileshch99589 @ubiOfficial @Cyberdost So much paper work, bhaag daur, why

Why can't we have a centralised system for issue like this. 🥲🚨

[AIA @000ADIL]

@ChapraZila Mera bihar badal rha hai, if it's true 🕵️‍♂️

[AIA @000ADIL]

@uttam_singhk Hypocrisy?? 😅🕵️‍♂️

[Programmer Humor @PR0GRAMMERHUM0R]

breakTheViciousCircle

[AIA @000ADIL]

"How does chunking work in your RAG pipeline?"

The problem first:

A PDF page has 2000 characters. If I pass the full page to the LLM — it gets 15 prompts when it only needs 1. Noisy context, higher hallucination risk, wasted tokens.

So I chunk it down.

How it works:

Sliding window across the text:

chunk_size = 500 chars
chunk_overlap = 100 chars
step = 400 chars (500 - 100)

Chunk 1: chars 0 → 500
Chunk 2: chars 400 → 900 ← overlaps by 100
Chunk 3: chars 800 → 1300 ← overlaps by 100

Why overlap?

Answers often sit at chunk boundaries. Without overlap you split a sentence in half and retrieval misses it.

Two details that matter in production:

Word boundary fix — never cut mid-word. Find the last space before the chunk end, slice there.

Metadata on every chunk — every chunk carries source, page_number, chunk_index.

When retrieval returns a wrong answer, I trace it back to the exact page instantly.

Why not just use LangChain's splitter?

It works, but returns plain strings — no metadata.

In production when something breaks, you need to know where that chunk came from.

Plain strings give you nothing to debug with.

#debugging
#pipeline
#AI

[AIA @000ADIL]

You cannot debug what you cannot trace.

#observability

[Indian Tech & Infra @IndianTechGuide]

🚨 A Kerala man applied for a government job in 2005 and has received an appointment letter as a part-time junior Arabic teacher in 2026. 20 years later. 🙏

[AIA @000ADIL]

python-dotenv → secrets safe
pydantic → data safe
ollama → local LLM, zero cost
langchain → document + RAG plumbing
PyMuPDF → clean PDF text
rank-bm25 → keyword search
sentence-transformers → local embeddings + reranking
chromadb → dev vector DB
qdrant-client → prod vector DB
langgraph → agent brain
ragas → proof it works
fastapi → expose it to the world
rich + loguru → see what's happening

#LLM
#AI
#Engineering
#Pipeline
#RAGA
#Learning
#langgraph
#Python

[AIA @000ADIL]

is it True?🧐🧐🧐

If I were optimizing for a job in 2026

I would honestly do:

Current RAG Project
↓
RAGAS
↓
Qdrant
↓
LangGraph
↓
Build One Serious Agent
↓
Deploy
↓
Observability

and stop there.

That portfolio is stronger than:

10 LangChain tutorials
50 prompt engineering projects

combined.

#AI
#Engineering
#LLM
#Learning
#Jobrelevant

[AIA @000ADIL]

Before coding, understand the core LangGraph idea:

State = shared data moving through the graph

Node = function that reads/updates state

Edge = path from one node to another

Conditional edge = branch based on state

#llm
#langgraph
#AI
#Engineering
#Production
#RealWorld
#Langchain

[AIA @000ADIL]

🔍 BM25 vs. Vector Search: What is the difference?

If you are building an AI search engine, you will hear about BM25 and Vector Search.

What do they actually do?

Here is the simple breakdown.

1️⃣ BM25 = The Keyword Matcher

BM25 is a traditional, keyword-based search.

It does not understand "meaning." It only looks for exact words.
How it scores things:

Did you use the exact word?

How many times did you use it?
How rare is that word in the whole database? (Rare words score higher than common words like "the").
Example:

You search: "car repair manual"

Document A says: "This is a car repair manual." -> High BM25 Score (Exact match!)

Document B says: "An automobile maintenance guide."

-> Zero BM25 Score (Different words, even though the meaning is identical).

2️⃣ Vector Search = The Meaning Matcher

Vector search is the modern, AI-powered way.

It converts text into numbers (called embeddings).

It does not care about exact words; it cares about the context and meaning.

How it scores things:

Are these concepts mathematically close to each other in meaning?

Example:

You search: "car repair manual"

Document B says: "An automobile maintenance guide."

-> High Vector Score (The AI knows "automobile" = "car" and "maintenance" = "repair").

Document C says: "The car repair shop is closed."

-> Lower Vector Score (It has the exact words, but the AI knows the context is about business hours, not a manual).

💡 Why we use both (Hybrid Search)

If Vector Search is so smart, why keep BM25?

Because Vector search can sometimes "hallucinate" meaning and miss exact part numbers or names.

BM25 is perfect for exact matches (like searching for a specific ID: "User_8472").

Vector Search is perfect for broad concepts (like searching "How do I reset my password?").

The Winning Strategy:

Run BM25. Run Vector Search.

Combine their lists using RRF (Reciprocal Rank Fusion) so you get the best of exact keywords AND deep meaning!

💡 The One-Line Interview Answer:

RRF combines search rankings without trusting incompatible raw scores.

#LLM
#Engineering
#AI
#RRF
#AIengineering

[AIA @000ADIL]

Vector score = distance, lower is better

BM25 score = keyword score, higher is better

#llm
#engineering
#AI
#AIEngineering

[AIA @000ADIL]

@piyushgarg_dev Bro just wanna know, why git is still alive
??

[Jahir Sheikh @jahirsheikh8]

POV: you joined an early-stage startup

[AIA @000ADIL]

We improve candidate recall by adding true BM25 search over all chunks.

Vector-only search = semantic candidates

BM25 over vector candidates = reranking only

True hybrid search = vector candidates + keyword candidates

#llmengineering
#aiengineering
#ai
#bm25
#hybridsearch

[AIA @000ADIL]

Current mood 🧐

PDF → Chunks → Embeddings → ChromaDB → Retrieval → Ollama → Answer

All local.
No OpenAI.
No Anthropic.
No external APIs.

Just learning how production-grade RAG systems actually work.

LLM Engineering journey has officially begun 🚀

#AIEngineering #RAG #Ollama #Python