public class AtharShaikh {
private final String role = "Software Developer (Backend + AI)";
private final int experience = 4; // years
private final String location = "Ahmedabad, India 🇮🇳";
private final String[] domains = {"Healthcare", "FinTech", "Telecom"};
public String mission() {
return "Turn complex business problems into fast, reliable, scalable systems.";
}
}I'm a Software Developer with 4 years of experience building scalable web applications, microservices, and REST APIs, alongside shipping production GenAI features (RAG, LangChain, agentic workflows).
I like systems that stay fast under load, stay readable six months later, and make the on-call Developer's life easy.
- 🏗️ Designing event-driven microservices that scale
- 🤖 Building RAG pipelines and agentic workflows with LangChain and LangGraph
- ⚡ Tuning PostgreSQL queries, caching and API latency
- 📚 Always learning: system design, distributed systems, AI Developering
A typical event-driven setup I build:
flowchart LR
Client([Client / UI]) --> GW[API Gateway]
GW --> A[Auth Service<br/>Spring Security]
GW --> B[Core Service<br/>Spring Boot]
B -->|events| K{{Kafka}}
K --> C[Notification Service]
K --> D[AI Service<br/>FastAPI + LangChain]
B --> DB[(PostgreSQL)]
B --> R[(Redis Cache)]
D --> V[(Vector Store)]
D --> LLM[[LLM]]
- Loose coupling: services talk through events, not tight calls
- Cache smartly: Redis for hot reads, PostgreSQL as the source of truth
- Secure by default: auth and validation at the edge, least privilege inside
- Observable: if I can't measure it, I can't fix it
I don't just call an LLM API. I build the full pipeline around it:
| Area | What I do |
|---|---|
| RAG | Chunking, embeddings, retrieval, and grounded answers over private data |
| Agents | Multi-step agentic workflows with LangChain and LangGraph |
| Integration | Wrapping AI features in production-grade Java and Python services |
| Reliability | Guardrails, fallbacks and sensible latency and cost control |
Most of my production work is for clients under NDA, so the source isn't public. The repositories below show how I design and build systems.
| Project | What it does | Stack |
|---|---|---|
| Scalable E-Commerce Backend | Microservices architecture with event sourcing, saga pattern, and distributed caching | Java, Spring Boot, PostgreSQL |
| Enterprise GenAI Knowledge Agent | Hybrid search RAG assistant with conversational memory and agentic tool routing | Python, FastAPI, LangChain |
| Event-Driven Stream Processing | High-throughput message streaming with fault-tolerant delivery and real-time synchronization | Kafka, Redis, Docker |
- 🔹 Advanced agent orchestration with LangGraph
- 🔹 Distributed systems patterns (saga, outbox, CQRS)
- 🔹 Performance Developering and observability
I'm happy to talk about backend architecture, GenAI in production, or interesting Developering problems.
"Make it work, make it right, make it fast." ⚡