Available commands: about - Who I am education - My academic background experience - My professional work history skills - What I can do projects - Stuff I've built achievements - Competitive programming stats contact - How to reach me

Hey, I'm Arnav Patil. I am a Software Engineer at Cyware Labs building AI-powered systems for threat intelligence at scale. My work spans NLP search engines, database-configurable AI summarization, and high-throughput data pipelines processing millions of objects daily. I care about system design, reliability under multi-tenant load, and making tools accessible to non-technical users. I build with Python, Django, FastAPI, Kafka, and AWS.

Indian Institute of Information Technology, Nagpur (2021 - 2025) - Bachelor of Technology in Computer Science and Engineering - CGPA: 8.79

Software Engineer - Cyware Labs India Pvt. Ltd (Sep 2025 - Present) - Built an AI-powered NLP search engine converting plain-English queries into complex CQL across 100+ fields with nested boolean logic, letting non-technical analysts at Fortune 500 and ISAC clients run threat hunts at 3–4s latency. - Architected a database-configurable AI summarization system, extending coverage from 2 to 6 object types (25+ indicator subtypes) with zero code changes to onboard new types; design doc adopted as team standard. - Engineered a threat-intelligence aggregation pipeline with two-level deep graph traversal processing millions of objects daily, cutting database queries 60% through a multi-layer caching strategy. - Closed a platform-wide unbounded-query vulnerability with a pagination guardrail (500-record cap) across 6 microservices, reducing worst-case response payloads up to 20× and stabilizing PostgreSQL/Elasticsearch under multi-tenant traffic. - Primary responder for a P0 outage (API-gateway failure + database corruption), restoring customer access within SLA; resolved 50+ escalations and 100+ production bugs across AWS services. - Cut CI pipeline runtime 50% (60→30 min) and unit-test latency 99% (600s→6s) by profiling and refactoring the test suite. Software Development Engineer Intern - Cyware Labs India Pvt. Ltd (Aug 2024 - Aug 2025) - Designed a STIX custom object deduplication system across 2 microservices, eliminating 70% of redundant ingestions and reducing Postgres/Elasticsearch storage by 40% while cutting code complexity by 40%. - Shipped 5+ backend features using Django, Kafka, AWS, and Elasticsearch for enterprise data ingestion pipelines, collaborating cross-functionally with DevOps and frontend teams. - Resolved 100+ bugs across customer-facing portals and data ingestion services, addressing configuration, validation, and parsing issues in production AWS environments.

Languages: - C, C++, Python, Java, JavaScript Databases: - PostgreSQL, Redis, Elasticsearch Web Frameworks: - Django, FastAPI Concepts: - REST APIs, Microservices, Event-Driven Architecture, Caching Developer Tools: - Git, Docker, Kafka, AWS

1. PyLSM: LSM-Tree Storage Engine - Tech: Python, FastAPI - Built a persistent key-value storage engine from scratch on the LSM-Tree architecture (RocksDB/Cassandra/Bigtable style): custom binary SSTable format (versioned, CRC32-validated), fsync-durable Write-Ahead Log, 7-level leveled compaction, and a FastAPI interface — zero external database libraries. - Reduced worst-case read I/O by ~94% with a per-SSTable Bloom filter (double-hashing, 1% false-positive rate) paired with a sparse index, replacing full-file linear scans with binary search over a bounded block. - Enabled lock-contention-free concurrent reads alongside a background compaction daemon (k-way min-heap merge) via a lock-snapshot pattern where readers copy the file manifest under a mutex and release before all disk I/O. - Validated with 77 passing tests across 7 modules; benchmarked against SQLite (WAL mode, 50K ops): ~24× faster random reads, 9× sequential reads, 1.45× sequential writes.

Competitive Programming: - Global Rank 66: CodeChef Starters 116 (Div 3) - Active participant with a consistent problem-solving history across CodeForces and CodeChef.

You can find me here: Email: arnavnpatil@gmail.com GitHub: arnav029 LinkedIn: arnav-patil