Adnan Shakil

Backend & Full-Stack Engineer · ML Systems Integration

Bengaluru, Karnataka, India

@adnan_shakil

GitHub LinkedIn

0 followers

🎓 B.Tech in Computer Science & Engineering at Presidency University, Bengaluru · Graduating 2028

About

Second-year Computer Science Engineering student with hands-on experience building and deploying full-stack AI-powered platforms, integrating Spring Boot/Node.js backends with Python ML pipelines, multi-agent systems, agentic RAG, and authentication from scratch. Recently completed a software engineering internship focused on production serverless infrastructure on AWS.

Experience

Software Engineering Intern — Backend & AI Systems

Dyslexia Reading Tutor AI

Jun 2026 – Aug 2026

Evaluated voice-AI architectures and selected ElevenLabs for production. Designed and deployed a serverless event-driven pipeline on AWS, built Lambda functions for outbound call triggering, conversation polling, and HubSpot CRM sync, and architected the middleware orchestration layer for the system.

Education

B.Tech

Presidency University, Bengaluru

Computer Science & Engineering · 2024 – 2028

Skills

<3 months

C CSS JWT Oop HTML Java React Ragas Python FastAPI Node.js XGBoost Pgvector LangGraph Express.js JavaScript Google Adk cockroachdb Tailwind CSS scikit-learn Random Forest API Integration Understanding Data Structures and Algorithms

Tools / apps / platforms

3 to 6 months

Lambda Deepgram ElevenLabs EventBridge HubSpot software Amazon Web Services AWS

<3 months

Git Groq Gemini GitHub Render VS Code PostgreSQL Google Gmail GitHub Actions JetBrains IntelliJ IDEA

Projects

FakeOut AI — Voice Deepfake Detection System ↗

Python, XGBoost, Random Forest, wav2vec2, Librosa, FastAPI, React

Dual-model ML ensemble for classifying audio as real or AI-generated using wav2vec2 embeddings and acoustic features.
Cypher — Autonomous Multi-Agent Job Intelligence System ↗

Python, LangGraph, FastAPI, PostgreSQL, React, Groq, Gmail API, GitHub Actions

Autonomous 5-agent system with persistent memory, batched LLM scoring, and automated reply detection/outreach delivery.
bengaluru-commute-agent — Multi-Agent Commute Planning AI ↗

Python, Google ADK, Gemini 2.5 Flash-Lite, FastAPI, React, OpenRouteService

Multi-agent commute planning system with strict role separation, live congestion reasoning, and honest fallback behavior.
Pitwall — Agentic RAG System for F1 Race Intelligence ↗

Python, LangGraph, FastAPI, pgvector, PostgreSQL, Groq, RAGAS, React

Agentic RAG system that evaluates retrieval sufficiency before answering and uses pgvector similarity search over F1 data.
DeltaBox — AI-Powered Formula 1 Intelligence Platform ↗

Spring Boot, React, PostgreSQL, Python, XGBoost, Random Forest, JWT, Render

Full-stack AI platform for Formula 1 intelligence with ML predictions, what-if simulations, and custom authentication/RBAC.

🏆 Achievements & awards

Ranked 144th at Google AI Agent Builder Series 2026 · 2026

HiDevs × Google for Developers.
Top score in first technical evaluation round · 2026

FusionX Hackathon 2026 for FakeOut AI.
5th place, Build with TRAE Hackathon 2026 · 2026

CentSight AI financial decision simulator.
PR #3507 merged to topoteretes/cognee · 2026

OSS contribution for mem0→Cognee migration example.

Explore the Broxer community

💬 Q&A Ask about roles, salaries and interviews — and help others. 👥 Circles Join discussions on skills, companies and industries. 📰 Feed Follow updates and insights from the community.