AI Engineer|Data Enthusiast| Generative AI,LLMs & Data |Deep learning->Artificial Neural Nets |
Building production-grade AI systems by combining scalable data engineering, MLOps, and LLMOps practices to transform raw data into reliable, observable, and commercially impactful intelligence.
Software Engineer with 2 years of experience architecting production LLM applications,Data Engineering, and enterprise RAG pipelines. Specialized in LLMOps, Generative AI, and scalable data infrastructure.
Core Expertise:
- Production Data Engineering (Spark, Kafka, ETL at Scale)
- RAG Architectures (Vector DBs, Hybrid Retrieval, Evaluation)
- Real-Time AI Pipelines (Streaming, Low-Latency Inference)
- LLMOps (Prompt Versioning, Cost Tracking, Observability)
Voice-native document Q&A using Gemini 2.0 Multimodal Live API
- Sub-200ms latency with streaming audio responses
- ElevenLabs voice synthesis + Firestore vector search
- Datadog observability for TTFT & token burn tracking
- Stack: Gemini 2.0, Firestore, FastAPI, React, ElevenLabs
Enterprise-grade streaming pipeline for stock sentiment analysis
- Processes 1000+ news articles/day via Kafka + Spark Streaming
- 87% sentiment accuracy using FinBERT on financial news
- Live dashboard correlating stock prices with sentiment trends
- Stack: Kafka, PySpark, HuggingFace, PostgreSQL, Streamlit, AWS
Automated resume analysis with semantic matching
- Skills gap analysis & upskilling recommendations
- Semantic similarity scoring using embeddings
- PDF parsing + LLM reasoning pipeline
- Stack: Gemini 1.5 Flash, Python, PyPDF2, FastAPI
18 Deep-Dive Blog Posts covering production AI systems: Go check it ---
- [MCP Deep Dive: Universal Connector for LLMs]
- [NVIDIA Triton Servers for Production Inference]
- [LLM Evaluation: Accuracy, Latency, Performance]
- [Fixing LLM Bottlenecks with Custom CUDA Kernels]
- [Finetune LLMs 2-5x Faster with Unsloth]
- [Training on 1TB Datasets with 3GB RAM]
- [Custom Keras Data Generators]
AI & LLM: LangChain • CrewAI • Swarm • LangGraph • HuggingFace Transformers Gemini 2.0 • GPT-4 • FinBERT FAISS • Pinecone • ChromaDB • Firestore Vector Search
Data Engineering: PySpark • Apache Kafka • Airflow • Great Expectations PostgreSQL • Redis • DynamoDB • Redshift AWS (S3, Lambda, Kinesis, EC2) • GCP (Cloud Run, Firestore)
Backend & APIs:
Python (FastAPI, Flask) • Node.js • TypeScript Docker • GitHub Actions • CI/CD
Open to:
- 🤝 Collaboration on AI/LLM projects
- 💼 AI Engineer / Data Engineer roles