Backend & Systems Software Engineer specializing in distributed architectures, high-throughput pipelines, API design, and database query optimization.

I focus on building production-oriented backend software with an emphasis on low-latency data access, background queue processing, multi-layer caching, and microservices design.

• Personal Hub: https://allkvd.dev/
• Portfolio: https://portfolio.allkvd.dev/
• Resume: https://drive.google.com/file/d/1NCT6ZCa_HfxCdScqI-1Q2yA6y2c7O-qA/view
• LinkedIn: https://www.linkedin.com/in/venkata-dhanush-k/
• X: https://x.com/kvdhanush06
• YouTube: https://www.youtube.com/@kvdhanush0608
• Medium: https://medium.com/@kvdhanush06
• Stack Overflow: https://stackoverflow.com/users/30895562/venkata-dhanush-kakarlamudi

• GATE 2026 CS Qualified — Computer Science & Information Technology
• Software Engineer Intern experience at StudyCubs & AK Capital Advisors
• Open Source Contributor to PrefectHQ/fastmcp
• Smart India Hackathon 2025 — National Round Qualifier

• Languages: Python • TypeScript • JavaScript • SQL • Java • C/C++
• Frameworks & Libraries: FastAPI • Express.js • Django • React • SQLAlchemy • Alembic • Pydantic
• Databases & Messaging: PostgreSQL • SQLite • Redis • RabbitMQ • BullMQ • Supabase
• Cloud & Infrastructure: AWS • OCI • Docker • Nginx
• Developer Tools: Git • GitHub Actions • Postman

• Architected a high-throughput data platform for 5K+ users, offloading real-time webhooks and 180-day batch syncs via RabbitMQ across 6 parallel worker streams.
• Reduced database lookup complexity from $O(N)$ table scans to $O(\log N)$ index seeks using a composite B-tree index on user and date fields for atomic upserts.
• Engineered fault-tolerant ingestion pipelines via 500-item chunked bulk upserts and nested transactions to isolate data failures and prevent full batch rollbacks.

• Designed 30+ credit and loan REST APIs, implementing strict financial domain validation via Pydantic schemas and OAuth2/JWT authentication dependencies.
• Executed 9 database schema migrations across 15 relational tables and 2 views using Alembic, maintaining strict data integrity via foreign keys and unique constraints.
• Standardized API error semantics by deploying a centralized error catalog with explicit transactional rollbacks to preserve database state during failures.

Framework for building Model Context Protocol (MCP) servers and clients in Python.

• PR #3736 — Patched an asynchronous lifecycle state leak by dynamically provisioning a fresh session manager per runtime cycle.
• PR #3750 — Hardened token validation using thread-safe deep-copy model replication for identity-claim propagation.
• PR #3768 — Resolved an OpenAPI-to-JSON Schema nullable/enum mismatch affecting autogenerated clients.

• Distributed retrieval platform coordinating external services and queue-backed workers with retries, Redis caching, citation validation, and streamed AI responses.
• Tech Stack: TypeScript, Node.js, Express.js, Redis, BullMQ, PostgreSQL, Supabase, React, Docker, Nginx
• Live Demo • Repository

• Django application combining study management, learning-resource discovery, utility tools, and workflow automation.
• Tech Stack: Django, Python, SQLite, Requests, Bootstrap
• Live Demo • Repository

• Django social platform focused on authenticated sessions, profiles, posts, social relationships, discovery, and relational feed retrieval.
• Tech Stack: Django, Python, SQLite, HTML/CSS, Pillow
• Live Demo • Repository

• Browser extension and web analyzer that parses Terms of Service and Privacy Policies to surface actionable risk insights and clause breakdowns.
• Live Demo • Repository

• Searchable directory for discovering AI applications by category, use case, pricing, and practical guides.
• Live Site • Repository

• Browser-based device-to-device transfer platform for sending files, folders, and text without uploads, accounts, or cloud storage, using WebRTC DataChannels for direct peer transfer and a lightweight WebSocket signaling server.
• Tech Stack: React, TypeScript, Node.js, WebSocket, WebRTC, Vite, Docker
• Live Demo • Repository

AllKVD is the canonical hub for my personal engineering work, products, portfolio, and professional links: https://allkvd.dev/

For current professional information, use the Venkata Dhanush Kakarlamudi portfolio and resume.

I am actively seeking Software Engineering, Backend Engineering, and SDE-1 opportunities where I can build scalable microservices, high-throughput APIs, and distributed architectures.

• Email: kvdhanush06@gmail.com
• LinkedIn: linkedin.com/in/venkata-dhanush-k
• Personal Hub: allkvd.dev