Open to Junior ML Eng roles

Backend Engineer,
ML Systems Builder.

I build production-grade backend systems and deploy machine learning pipelines — Java & Spring Boot depth, Python for ML/AI, now extended into LLMs, RAG, and real-time ML infrastructure.

Get in touch View work Resume
● Available · IST
Sonu
Backend & ML Engineer
BTech CS · Manav Rachna · 2026
Java 21 Spring Boot 3.5 Python FastAPI PyTorch pgvector RAG Docker RabbitMQ PostgreSQL

Academic
Background

Computer Science graduate with production experience across backend systems, ML pipelines, and LLM infrastructure.

B.Tech Computer Science
Manav Rachna University · Faridabad
2022 — May 2026

Graduated May 2026. Core focus on software engineering, data structures, and distributed systems. Supplemented academics with ~3 years of Java/Spring Boot production experience and an intensive self-directed ML engineering curriculum covering classical ML, deep learning, LLMs, RAG, MLOps, and production system design.

Java Python Spring Boot Distributed Systems

Things I've Built

GitHub

LexGuard

Legal document ingestion & RAG platform. Async Spring Boot 3.5/Java 21 service with Transactional Outbox, RabbitMQ, Cloudflare R2, and pgvector HNSW (m=16). Python FastAPI RAG layer with SHA-256 auth, SlowAPI rate limiting, prompt injection defense, Prometheus + Grafana observability.

● Phase 1 complete · 6-service Docker stack
Java 21 Spring Boot 3.5 Python FastAPI pgvector RabbitMQ Docker
GitHub Live

LLM Cost Router

3-layer intelligent routing pipeline: SemanticCache (FAISS + Redis, 87% hit rate) → TF-IDF + LogReg classifier → LLM fallback. FastAPI with /route, /benchmark, /health endpoints. Deployed on Render.

● 93.19% cost reduction · 100% routing accuracy
Python FastAPI FAISS Redis scikit-learn Docker
GitHub Live

Financial RAG API

Hybrid BM25 + dense vector search over Apple & Microsoft 10-K filings. Docling ingestion (74→453 chunks), dynamic ChromaDB collections, Groq LLaMA-3.3-70B generation, MLflow + Prometheus observability. Gradio UI with company selector.

● Context Recall 0→1.0 post-Docling · Faithfulness 0.82
Python FastAPI ChromaDB Groq MLflow Docker

Tools of
the Trade

Production-tested backend depth, now extended into ML infrastructure and LLM systems.

01
Programming Languages
Java Primary
Java 21
Java 21 Spring Boot 3.5 / 4 OOP & Design Patterns Concurrency Generics JVM Internals
Python Primary
ML / Backend
FastAPI PyTorch scikit-learn Pandas / NumPy asyncio Pydantic SQLAlchemy
02
Backend Engineering
Spring Boot 3.5 Spring Boot 4 FastAPI REST APIs JWT WebSockets AOP Transactional Outbox
03
ML & AI Systems
PyTorch Transformers RAG LLMs (Groq / Gemini) ChromaDB pgvector FAISS SHAP LoRA (from scratch) INT8 Quantization Drift Detection
04
Data & Messaging
PostgreSQL Redis RabbitMQ Kafka Flyway SQLAlchemy Airflow
05
Infra & Observability
Docker Docker Compose Cloudflare R2 Render Prometheus Grafana MLflow DVC

Let's build
something together.

Open to new opportunities in backend and ML engineering. If you're building something interesting, I'd love to hear about it.

Local time
—
—
Typical response
Within 24 hours
Open to
Full-time · Contract · Freelance
Work preference
Remote · All timezones
Currently building
LexGuard · ML Engineering curriculum
Drop me a line