Pratham Rao

I’m a B.Tech student in Computer Science and Engineering at PES University, graduating in June 2026, with a GPA of 9.39/10 (Top 5% of my department). I enjoy turning complex AI workflows into reliable production-grade systems, especially where latency, throughput, and cost matter.

As a Software Development Intern at a stealth startup (AI-native B2B marketing platform), I built v3 of an on-demand content generation workflow using an async Temporal pipeline and Java→Python child workflows. I re-engineered an agentic screenshot tool (Playwright + sharp) to cut per-image size and LLM token cost by ~13×, migrated validation into a versioned Spring Boot service, and applied a claim-check pattern to stay within Temporal’s 2 MB payload limit.

I also shipped core multi-agent tooling (Next.js + Mastra) for conversation flows, including content-type-aware tools and automatic Anthropic→Vertex failover across model providers. Alongside building distributed logging and Redis-compatible infrastructure in projects, I’ve contributed as a Teaching Assistant for Big Data and as a Research Intern where my work on darknet-marketplace data led to an ICSPIS 2024 (IEEE) paper.