I’m a Software Engineer with a strong focus on data systems, backend engineering, and production AI/ML systems. I recently completed my MS in Computer Science at Indiana University Bloomington, where I combined solid CS fundamentals with hands-on experience building scalable, real-world systems.

I enjoy working close to the system boundary — thinking about performance, reliability, evaluation, and tradeoffs — and shipping solutions that hold up in production.

• Data & Platform Engineering Designing and deploying Spark / Databricks pipelines, ingestion workflows, caching layers, and backend services with clear throughput, cost, and fault-tolerance goals.

• Production AI & ML Systems Building NLP and ML systems end to end — from training and evaluation to deployment, monitoring, and iteration.

• Retrieval & Search Systems Developing RAG pipelines, vector indexing, semantic search, and ranking systems over large document corpora.

• Backend & API Development Shipping FastAPI / Flask services with batching, caching, observability, and clean interfaces for downstream users.

• Built a retrieval-augmented QA system over 10K+ documents
• Improved Recall@5 from 0.62 → 0.87 using HNSW indexing and cross-encoder reranking
• Deployed with FastAPI, Redis caching, and Docker
• Focused on latency, memory efficiency, and cost optimization

• Designed a tool-using agent with a structured plan → act → observe → reflect loop
• Achieved 19/20 task success with validation, retries, and timeouts
• Published to PyPI with logging, tracing, and documented failure modes

• Trained Transformer models on 50K sentence pairs
• Improved BLEU score from 28.4 → 32.7
• Implemented beam search decoding and deployed an inference API

• Software Engineer – ML & Data Systems (Contract) Built Spark SQL ingestion pipelines on Databricks, evaluation frameworks, and batch / near-real-time workflows with caching and idempotent writes.

• Computer Vision Research Assistant Trained and deployed 3D point cloud models (PointNet++, KPConv), built annotation tooling, and shipped inference APIs for large-scale datasets.

• Software Engineer – ML Systems Developed multimodal NLP systems, fine-tuned large models, and engineered data pipelines processing 100K+ records.

Languages & Frameworks Python, SQL, FastAPI, Flask, Django, JavaScript

Data & Distributed Systems Spark (PySpark, Spark SQL), Databricks, Data Pipelines, Distributed Processing, Redis

AI / ML PyTorch, TensorFlow, Transformers, scikit-learn, NLP, RAG, Model Evaluation

Production & Cloud Docker, Git, CI/CD, AWS, GCP, MLflow, Weights & Biases

• Software Engineering (Backend / Full Stack)
• Data, Platform, and Infrastructure roles
• Production-focused AI / ML Engineering

I’m especially excited by teams building systems, platforms, and developer-facing tools that scale to real users.

• Email: atharvagurav01@gmail.com
• LinkedIn: linkedin.com/in/atharvagurav01

Always open to conversations about engineering, systems, and building things that matter.