AI · 30 Sep 2026 · 6 min read
5 Security Checks Every App Builder Must Run Before Launch
A practical checklist of 5 AI-assisted security audits — secrets, data flow, production hardening, deep vulnerabilities, and the attacker's perspective.
Avaneesh Kumar Yadav · Engineering Manager
Building AI.
Scaling Systems.
Leading Engineering.
Engineering lessons from 12+ years in enterprise AI and cloud systems: how production AI is designed, run and led. Written as articles, courses and small tools you can use.
Articles
Technical writing on production AI
Architecture, trade-offs and what goes wrong, written for engineers who have to run these systems.
AI · 17 Sep 2026 · 10 min read
I Rebuilt Our AI Agents on Kafka Instead of REST — Here's the Full Java Architecture
REST between AI agents works until agent #4 shows up. The full Java/Spring Kafka topic design, the ReAct loop as a state machine, and a free audit trail.
AI Engineering · 9 Sep 2026 · 9 min read
Designing the AI Request Pipeline: 8 Layers Between User Input and Your LLM
A production architecture blueprint for every layer between a user's message and a safe, reliable LLM response — with Spring Boot code you can actually use.
Java Engineering · 9 Sep 2026 · 9 min read
The Modular Monolith: The Java Architecture Most Teams Should Be Using
A practical guide to the modular monolith in Java and Spring Boot — package structure, module boundaries, domain events, and why it beats microservices.
Architecture lab
Diagrams and decision frameworks
Reference shapes for production AI systems, and the questions that decide between the common options.
From request to response
Illustrative architectureFailure scenarios to design for: the retriever returns nothing relevant; the primary model is slow or unavailable; the output fails the schema or policy check; a prompt change quietly lowers quality. Each needs a defined behaviour, not a hope.
An agent you can stop
Illustrative architectureFailure scenarios to design for: the loop never terminates; a tool call has more privilege than the task needs; an irreversible action runs unreviewed; a wrong observation is trusted and compounds.
Retrieval, long context or fine-tuning?
A way to ask the right questions, not a benchmark. Your data, volume and constraints decide.
- The knowledge changes often, or is private to you.
- Retrieve it at request time. Keep the index easy to update, filter by who is asking, and show the sources.
- The whole body of knowledge is small, stable and fits comfortably in the context window.
- Try putting it in the prompt first. Add retrieval only when it stops fitting or the cost becomes a problem.
- You need a consistent format, tone or behaviour that prompting does not deliver.
- Consider fine-tuning, but only after an evaluation set shows that prompting and retrieval fall short.
- You cannot yet say how you would know it works.
- Build the evaluation set first. Every other choice here depends on it.
Go deeper
- Request pipeline in depth: Designing the AI Request Pipeline: 8 Layers Between User Input and Your LLM
- Memory for long-running agents: Agent Memory Patterns: How to Give Your LLM Application a Brain That Persists
- Shipping model changes safely: Feature Flags for AI: The Deployment Pattern That Saved Us From 3 Production Disasters
Tools
Small tools that do one job
Free calculators and advisors for AI engineering decisions. They run in your browser or call a small API.
AI Opportunity Finder
NewDiscover 3 personalized AI business opportunities based on your skills, experience, and goals — with market scores, a revenue roadmap, and a 30-day action plan.
LLM Cost Estimator
Estimate monthly spend from token volume and per-million-token pricing — across input and output.
RAG Chunk Calculator
Plan chunk size, overlap, chunk count, and one-time embedding cost for a document corpus.
Which Claude Model Should I Use?
Answer three quick questions and get a model-tier recommendation with the reasoning.
Live · Powered by Llama 3.1
Try the AI demo
Ask an enterprise-AI question and see a live, RAG-style answer — a small demo of the patterns I write about.
Click an example question or type your own above.
Production-grade answers in seconds.
Powered by Groq · Llama 3.1-8B-Instant · A live demo, not professional advice
Learning instead? Free courses and Claude certification practice exams.
About
Avaneesh Kumar Yadav
Engineering Manager
12+ years in enterprise AI and cloud systems. I write about how production AI is designed, run and led, and turn it into free courses and small tools.