.png){kind=logo}
SP
Sai Prakash Sasubilli
Mid-level Full-Stack Software Engineer specializing in cloud-native web apps and AI agents
NH, USASoftware Engineer (Remote)4 years experienceMid-LevelTechnologySaaSWeb Development
ScreenedIdentity Verified
Connect with Sai
Sai already has a relationship with Reval, so a warm intro from us gets a much better response than cold outreach.
Recommended
Already have an account?
About
Full-stack system analyst/programmer at PeakPlay Sports (startup) who built an AI "coach" product end-to-end in ~2 months, using a LangGraph-orchestrated multi-agent architecture with a FastAPI backend. Shipped production RAG grounded in athlete history (OpenAI embeddings + vector store) with guardrails and a structured eval loop (golden set + LLM-judge + human review) to improve engagement and reduce hallucinations.
Experience
Software Engineer (Remote)Peak Play AI Sports
Graduate Teaching AssistantRivier University
Software EngineerMagic Software
Education
Rivier Universitymaster, Computer Information Systems
Acharya Nagarjuna University – Bapatla Engineering Collegebachelor, Electronics and Communication Engineering
Key Strengths
- Owned end-to-end build from architecture through deployment in a startup environment
- Designed multi-agent LLM coaching system with parallel orchestration and conditional execution
- Balanced latency vs response cohesion; improved synthesis via head-coach prompt/architecture changes
- Improved user engagement by restructuring outputs into prioritized action items and table-based plans
- Built production retrieval layer grounding outputs in athlete history using embeddings + vector search
- Implemented LLM guardrails (relevance thresholding, source-attribution prompting) to reduce hallucinations
- Created evaluation loop with curated golden set (40–50 cases), automated scoring, and weekly human review
- Used eval findings to prevent regressions (ensured synthesis includes points from each coach, incl. biomechanics)
- Built normalization and reliability layer for messy real-world inputs with graceful failure and retries
- Implemented observability to trace raw vs normalized inputs, node execution, and report success/failure
Discover more candidates like Sai
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.
Search Talent