Open to new opportunities
Jerusan Jegatheeswaran

Jerusan Jegatheeswaran

I've shipped across fintech, compliance, and legaltech - from trade execution engines to SOC 2 automation. Full-stack enough to own things end-to-end, particular about the parts no one sees.

LegalTech Fintech GRC

Experience

2026
Present

ComplianceOS

Founder // Minneapolis, MN
  • Architected a SOC 2 automation platform designed to significantly reduce manual audit workflows via automated evidence collection.
  • Engineered control evaluation pipelines reducing audit preparation time from days to a few hours.
  • Implemented real-time exception tracking across multi-cloud environments.
2022
2023

Midpoint Labs

Software Engineer // SF
  • Reduced annual infrastructure spend by $12K by building a custom blockchain listener.
  • Built an event-driven Go backend processing 50K+ daily actions with sub-second latency using distributed queues and real-time aggregation pipelines.
  • Achieved a 65% reduction in production downtime through proactive health monitoring, custom alerting logic, and AWS infrastructure integration.
2022

Diffuse Funds

Software Engineer // Chicago
  • Developed a trade engine handling 1,000 TPS, optimizing for block inclusion priority.
  • Built a TypeScript metadata API aggregating smart contract ABIs across 20+ DeFi protocols, cutting new protocol integration time by 80%.
  • Automated reconciliation pipelines in Python, compressing workflows from days to ~10 minutes.
2019
2022

KLDiscovery

Software Engineer // MN
  • Redesigned RBAC systems in C#, reducing support tickets by ~70%.
  • Architected an automated case onboarding workflow that eliminated manual storage allocation bottlenecks for document-heavy legal workflows.
  • Built a multithreaded analytics engine processing millions of records monthly.

Featured Work

ASVec

C++SIMDHNSW
489 QPS at 98.95% recall@100 on SIFT1M
7.5× throughput gain over exact SIMD search
speedup over scalar path via ARM NEON

HNSW approximate nearest neighbor search in C++ with SIMD vectorization, SoA memory layout, and concurrent graph construction across all hardware threads.

[ source_code ]

Dex

Next.jsPythonLLMs
<1s sandboxed code execution latency
horizontal scale via CloudFront

Voice-based AI interviewer for multi-stage technical assessments with real-time LLM feedback on problem-solving, algorithm design, and code quality.

[ source_code ]

Tech Stack

C++ Go Python TypeScript Node.js C# PostgreSQL React
Distributed Systems Event-Driven Architecture Real-Time Pipelines High-Performance Backends Fintech GRC Tooling

Contact

Let's build something that scales.

I'm open to roles that solves challenging and interesting problems.