👋 Hi, I'm Brijesh Kumar

Full‑Stack Engineer · ML Researcher · Systems Builder

MS Computer Science @ Arizona State University · 4.0 GPA · Tempe, AZ

📍 Tempe, AZ · ☎️ 623‑666‑2665 · Open to CPT / OPT · Actively seeking Summer 2026 internships

• Build production systems that stay up: sub‑150ms latency, 99.99% uptime, and 10K+ concurrent users.
• Ship full‑stack platforms with Java/Spring Boot, Python/FastAPI, React/Next.js, PostgreSQL, Redis, and AWS.
• Design ML & GenAI pipelines (RAG, forecasting, embeddings) on real‑world data at TB scale.
• Work across the stack: backend architecture, data pipelines, microservices, and frontend UX.

Machine Learning Research Aide · Jan 2025 – Present · Tempe, AZ

I work on large‑scale microbiome and clinical datasets to power real‑world healthcare decision‑making:

• Deployed RAG‑LLM pipelines with cosine‑similarity vector search on clinical + metagenomic embeddings, improving biomarker discovery precision by 21% across 3 disease domains.
• Built Python + SQL + Snowflake ETL on a 1.2TB microbiome dataset, with skew‑aware partitioning that cut LLM feature prep time by 65% on nightly AWS training pipelines.
• Designed hybrid ARIMA + Random Forest forecasting pipelines, reaching 92% accuracy and reducing clinician forecast error by 31% versus prior baselines.
• Automated EDA, hypothesis testing, and statistical modeling (Python/R) across 1.2TB clinical cohorts, surfacing 17 high‑signal gut taxa that shaped downstream feature engineering and LLM features.
• Built Power BI dashboards on ETL outputs with KPI layers, cutting reporting turnaround by 40%, queried daily by 30+ clinical research stakeholders.

Sep 2025 – Present · Tempe, AZ

• Built GenAI tutoring and content pipelines with AWS Bedrock + Amazon SageMaker, delivering adaptive tutoring, AI content recommendations, and quiz generation for 10K+ ASU Online learners, boosting engagement by 28%.
• Developed Student Enrollment Service in Java Spring Boot (registration, seat allocation, waitlists), reducing REST API latency by 42% through query optimization and profiling.
• Implemented Course Delivery & Notification Services with async Spring and AWS Lambda triggers on S3‑backed content pipelines, increasing throughput by 35% and delivering real‑time grade alerts and deadline reminders.
• Automated infra (EC2, IAM, VPC, Lambda) using Terraform, eliminating config drift and cutting provisioning time by 60%.
• Designed GitHub Actions CI/CD with tests, lint gates, and zero‑downtime deploys, improving release velocity by 2.1×.

Mar 2024 – Dec 2024 · India

• Built Dockerized FastAPI microservices for auth, enrollment, and assignments on AWS EC2, cutting deployment cycles by 45%.
• Scheduled Cron‑based automation for progress reports and deadline reminders, reducing manual ops by 40%.
• Secured APIs with JWT/OAuth2 + Pydantic validation, reducing invalid requests by 80% and improving data integrity.
• Designed PostgreSQL schemas with compound indexes and JSONB fields, improving query performance 3× and cutting p99 latency by 40%.

Mar 2023 – Mar 2024 · India

• Built 15+ React components for Stripe checkout (card input, 3DS, confirmation), increasing payment success rates by 30%.
• Implemented Redis caching on high‑traffic payment endpoints, reducing DB load by 60% and end‑to‑end latency by 40%.
• Designed GraphQL + REST APIs with batching, retries, and circuit breakers, improving inter‑service reliability by 42%.

Software Developer Intern · Oct 2022 – Feb 2023

• Developed serverless ETL pipelines with AWS Lambda, S3, RDS, improving processing efficiency by 35%.
• Provisioned EC2, IAM, and VPC via Terraform, cutting environment setup time by 50%.
• Shipped 10+ REST APIs and removed N+1 queries via indexed joins, cutting average API response time by 35%; awarded Best Intern.

A few projects that represent how I design, build, and ship systems end‑to‑end.

Stack: Python · FastAPI · LangChain · FAISS / Vector DB · AWS
Focus: Clinical QA and biomarker discovery support using retrieval‑augmented generation.

• Designed an end‑to‑end RAG pipeline over medical PDFs, guidelines, and structured cohorts to answer clinician‑style queries with grounded citations.
• Implemented chunking, embedding, reranking, and prompt strategies to reduce hallucinations and keep responses tied to validated sources.
• Integrated evaluation scripts and prompt sets to systematically compare retrieval strategies and prompt templates.

GitHub: TrustmedAi (Advanced RAG system)

Stack: Next.js / React · Python / FastAPI · AWS (Lambda, API Gateway, S3, RDS or DynamoDB)
Focus: Personal study assistant that organizes notes, generates quizzes, and explains concepts using GenAI.

• Built a full‑stack learning platform where students upload content and receive AI‑generated summaries, flashcards, and practice questions.
• Deployed backend services on AWS with API Gateway + Lambda and persistent storage, designing APIs for session management, content ingestion, and retrieval.
• Implemented authentication, role‑based access, and a responsive UI so the app feels like a polished SaaS product rather than a demo.

GitHub: StudySliceAI

Stack: Python · FastAPI · LangChain / OpenAI · Slack, GitHub, Notion APIs
Focus: “Open‑claw–style” agent that connects Slack, GitHub, and Notion to answer questions and automate workflows.

• Orchestrated multi‑tool agents that can read Slack threads, fetch GitHub issues/PRs, and query Notion docs to answer “what’s going on with X?” in one place.
• Implemented tool abstractions for each integration (Slack, GitHub, Notion) and an orchestration layer to route user intent to the right tools.
• Designed the system so teams can plug it into their existing Slack workspace and immediately get value without heavy setup.

GitHub: SlackAgent

Stack: Java · Spring Boot · AWS Bedrock · FAISS · Redis · Apache Kafka
Focus: AI‑enhanced contact manager and relationship graph for business networking.

• Built an event‑driven microservices architecture with Kafka, aggregating contact events across systems into a unified graph.
• Integrated AWS Bedrock + FAISS to embed contact metadata and surface high‑value relationships, improving contact matching accuracy by 70%.
• Tuned pipelines to deliver sub‑3 second network analysis queries at scale using Redis caching and optimized queries.

GitHub: Nexus

Stack: Python · FastAPI · CLIP · ResNet50 · AWS
Focus: Real‑time fashion recommendation using image embeddings and similarity search.

• Built a computer‑vision‑powered recommendation system that serves personalized outfit suggestions in production.
• Optimized for <150ms response times for 10K+ concurrent users with 99.99% uptime, using efficient batching and model serving on AWS.
• Improved user engagement by 32% with targeted recommendations versus simple rule‑based suggestions.

GitHub: FashionRecommender

Backend & Systems

• Languages: Python · Java · TypeScript · JavaScript
• Frameworks: Spring Boot · FastAPI · Flask · Node.js
• Architectures: Microservices · REST · GraphQL · Event‑Driven
• Messaging & Caching: Apache Kafka · Redis · Redis Pub/Sub
• Testing: Pytest · JUnit · Jest

Frontend & UX

• Frameworks: React · Next.js
• Styling & UI: Tailwind CSS · shadcn/ui · Material‑UI · Ant Design · Chakra UI
• Patterns: Responsive design, Progressive Web Apps, dashboard‑driven UX

AI/ML & Data

• Libraries: PyTorch · TensorFlow · Keras · scikit‑learn · Pandas · NumPy
• GenAI & NLP: Hugging Face · LangChain · RAG · Vector Embeddings
• CV: OpenCV · CLIP · ResNet
• Big Data: PySpark
• MLOps: MLflow · Weights & Biases
• Analytics & Viz: Power BI · Tableau · Matplotlib · D3.js

Cloud, DevOps & Infra

• Cloud: AWS · GCP · Vercel
• Orchestration & Containers: Docker · Kubernetes
• IaC & CI/CD: Terraform · GitHub Actions · Jenkins
• Monitoring: CloudWatch · Prometheus · Grafana · New Relic
• Databases: PostgreSQL · MongoDB · SQLite

╔═══════════════════════════════════════════════════════════════════════════════╗
║                         COMPETITIVE PROGRAMMING                              ║
╠═══════════════════════════════════════════════════════════════════════════════╣
║  🥇  Top 0.2% in CodeKaze — Rank 315 out of 150,000 participants            ║
║  🧩  1,000+ LeetCode Problems Solved (C++)                                   ║
║  🏆  Hackathon Champion (4×) — top 1% out of 10,000+ participants           ║
║  🎯  Finalist: JPMorgan Code for Good 2024                                  ║
╚═══════════════════════════════════════════════════════════════════════════════╝

╔═══════════════════════════════════════════════════════════════════════════════╗
║                        RESEARCH & CERTIFICATIONS                             ║
╠═══════════════════════════════════════════════════════════════════════════════╣
║  📄  Research Publication — "Early Dementia Detection via ANN Segmentation" ║
║      Published in Springer · Cited across academic repositories             ║
║  🎓  IBM Data Science Certificate (Coursera) — Top 5% globally              ║
║  🧠  Deep Learning Specialization (DeepLearning.AI)                          ║
║      CNNs · RNNs · NLP · Optimization for real-world AI systems             ║
║  🌍  Open Source — Hacktoberfest 2025 · 6+ merged PRs across OSS repos     ║
╚═══════════════════════════════════════════════════════════════════════════════╝

I’m actively seeking Summer 2026 internships (Backend · Full‑Stack · AI/ML).

• ✅ CPT/OPT eligible
• 📍 Tempe / Phoenix, AZ (open to relocation)
• 🎯 Interested in scalable systems, GenAI products, and data‑driven platforms that actually ship to users

If you’re building something ambitious and need someone who can own the path from data → models → backend → production, I’d love to chat.