Kishore Prashanth
************@*****.*** +1-623-***-**** LinkedIn GitHub Portfolio
EXPERIENCE
Associate Software Developer Mar 2026 – Present
Echopad Tempe, AZ
• Built a tiered in-process cache (cachetools TTLCache, 5-min–24-hr TTLs by volatility) with single-flight locking to prevent cache-miss stampedes on Snowflake queries
• Implemented multi-provider auth (MSAL for Entra ID, @react-oauth/google for Google) with hand-rolled JWT verification
(python-jose, RS256/JWKS+HS256) and single-active-session enforcement via a per-request nonce checked against a Cosmos DB fingerprint
• Built the credit-enforcement system on Cosmos DB conditional PATCH with atomic increments and filter predicates, falling back to a shared tenant pool on cap violations
• Built a payer-rate pipeline processing multi-terabyte gzipped JSON (up to 31TB for one payer) via streaming (ijson) parsing to avoid loading files into memory, on Azure Batch pools autoscaled by file size (1,587 sub-tasks, 50 nodes, ~15 hrs in one run), with quarantine/manifest quality gates and an incremental Snowflake load reconciled by row count
• Closed a fail-open access-control bypass across 4 read endpoints by replacing ad-hoc checks with a single fail-closed entitlement guard, and fixed a type-mismatch bug that silently let searches go unbilled
• Used GitHub Copilot and Claude daily for prototyping, refactors, and code review across the codebase AWS Cloud Developer Jul 2025 – Feb 2026
MindSpark Tempe, AZ
• Modeled a live/analytical data split on AWS: DynamoDB as source-of-truth for real-time state (sessions, participants, per- user counters), synced end-to-end through two chained DynamoDB-Streams-triggered Lambdas that aggregate scores and roll them up into Aurora Postgres via the RDS Data API
• Built 8 Python Lambdas behind a single API Gateway “router” endpoint dispatching by action field, each under its own least- privilege IAM role, with CloudWatch alarms on error rate/duration and prompt/model config pulled at runtime via AppConfig
• Built “Ari,” a hand-rolled orchestration agent (GPT-4.1, structured JSON output) for participant matching and agenda gener- ation, plus a profanity-detection model tuned to huddle chat content
• Owned deployment and post-launch troubleshooting, writing documentation (Lambda inventory, IAM fix writeups, config reference) so teammates could self-serve recurring issues IT Senior Associate Aug 2021 – May 2022
NTT Data Bengaluru, India
• Migrated a client-facing AngularJS application to Angular for Banco Sabadell, coordinating API contracts with intercontinental teams across time zones
• Optimized the migrated app for performance (5% improvement) and resolved recurring bugs, cutting downtime by 15%
• Built “Academia Nexus,” a Java enterprise application (CLI + Eclipse GUI) with role-based access control, securing 5,000+ users and cutting unauthorized access by 55%
PROJECTS
Sentinel Command — Real-Time Dispatch & Routing System Next.js, TypeScript, React, FastAPI, PostgreSQL, WebSockets
• Built a resource-dispatch algorithm ranking available field units by haversine distance plus capacity/condition penalties, auto-provisioning backup capacity on the fly when no matching unit type is available
• Built an AI classifier parsing free-text reports into structured tasks, plus a background process that auto-resolves completed tasks and tracks unit location/capacity/condition over time School Recommendation System Python, Flask, Scikit-learn, Docker
• Built a school-recommendation system over a 220-school Phoenix-area dataset (~40 features per school), with three independently-selectable similarity models: KNN nearest-neighbor search, Random Forest via per-tree label polling, and SVM via top-N class probabilities
• Exposed recommendations through a Flask REST API (search by name, by preferences, by city) with a static frontend, containerized for deployment (Docker, gunicorn)
TECHNICAL SKILLS
Languages: Python, JavaScript, TypeScript, Java, SQL, C++, HTML/CSS Frameworks & API Design: React.js, Next.js, Angular, Node.js/Express, FastAPI, Flask, REST APIs, WebSockets AI / LLM Dev: LLM Integration (OpenAI GPT-4.1), Structured/JSON-Mode Prompting, GitHub Copilot, Claude, Scikit-learn Cloud, DevOps & Data: AWS (Lambda, API Gateway, DynamoDB, RDS, CloudWatch, Secrets Manager), Azure (Cosmos DB, Blob Storage, Batch, App Service), PyArrow/Parquet, ijson, Docker, Terraform, CI/CD (GitHub Actions) Databases & Security: PostgreSQL, DynamoDB, Cosmos DB, Snowflake, JWT, OAuth 2.0, IAM EDUCATION
Arizona State University M.S. Information Technology — GPA 4.0/4.0 2023 – 2025 Visvesvaraya Technological University B.E. Computer Science & Engineering — GPA 3.7/4.0 2017 – 2021