I'm Athul Thulasidasan, a software engineer based in Boston, MA. I build backend systems for AI workloads, distributed data platforms, and developer-facing tools with a focus on latency, reliability, and clear system design.
I'm most interested in the part of software engineering where performance, correctness, and operations meet. The work I enjoy most is taking a system that feels noisy or fragile, understanding where the actual bottleneck is, and making it faster and more reliable without making it harder to maintain.
What I work on
- AI infrastructure and inference systems
- Distributed backend platforms and APIs
- Reliability, observability, and production tooling
- Data-intensive services built with Python, Go, TypeScript, AWS, Docker, Kubernetes, PostgreSQL, Kafka, Redis, Prometheus, and Grafana
Selected impact
- Reduced AI inference workflow latency by about 50% by autoscaling Python services on AWS with Docker and Kubernetes.
- Lowered integration failures by about 33% by standardizing OpenAPI contracts across five business units with feature-flagged rollouts.
- Improved mean time to detection by about 40% by building Prometheus and Grafana based observability.
- Increased large-scale data processing throughput by about 2.5x through vectorized Pandas pipelines.
- Improved release stability to about 99% successful deployments with automated testing and CI quality gates.
Background
- MS in Computer Science, Boston University, Sep 2024 - Jan 2026
- BTech in Computer Engineering, University of Mumbai, Aug 2019 - May 2023
Beyond day-to-day engineering
- Winner, GWiSE NEU Hackathon
- Led a 10-member team for Alegria across 13 workshops and 1,500+ attendees
- Technical speaker on TRPC, Rust, ML, Flask, and DevOps across 5 meetups and tech events
- Research publication: "Determination of Medicinal Leaf Properties", IJSREM, Apr 2023, DOI: 10.55041/IJSREM19036
Contact