I am a software engineer with 6+ years of experience across distributed backend systems, AWS serverless architecture, Android engineering, WebRTC-based real-time systems, APK instrumentation, and AI/data tooling. My work spans backend infrastructure, mobile platforms, cloud gaming, and developer tooling.

Location: Montréal, QC, Canada Profiles: GitHub · LinkedIn Languages: English (native or bilingual) · French (elementary)

Experience

Inagral

Software Engineer _Remote · Canada · Nov 2021 - Feb 2026

Worked across Android cloud gaming, AWS serverless backends, real-time streaming infrastructure, APK instrumentation / DRM tooling, monitoring, telemetry, and internal AI/data projects.

Serverless Virtual Device Orchestration Control Plane

Built a serverless control plane for pools of containerized Android and virtual devices used in session-based workloads.

  • Architected a hybrid ECS/Fargate + AWS Lambda system for low-latency routing and asynchronous lifecycle management.
  • Implemented dynamic pool provisioning and reconciliation using SQS, DynamoDB, and EventBridge.
  • Achieved 30%–50% compute savings through precise warm-pool provisioning and automated teardown.

Tech stack: Python, FastAPI, AWS Lambda, ECS/Fargate, SQS, DynamoDB, EventBridge, SAM, ALB, VPC Networking

Serverless Signaling Service for Session Negotiation and State Sync

Designed a scalable AWS-based signaling and real-time synchronization layer for cloud gaming sessions.

  • Built WebRTC session negotiation and state synchronization services with AWS Lambda, DynamoDB, and WebSocket APIs.
  • Helped migrate from a legacy EC2-based setup to a more cost-effective serverless model.
  • Reduced operational cost by roughly 70%–80%.

Tech stack: Node.js, Python, AWS Lambda, DynamoDB, WebSocket APIs, WebRTC

Distributed Streaming Platform and WebRTC Media Server

Worked on low-latency media and infrastructure for cloud gaming and real-time streaming.

  • Built distributed streaming components on AWS EC2 with Auto Scaling Groups.
  • Developed C++ WebRTC media server components using GStreamer.
  • Supported automated recovery and stable low-latency video delivery.

Tech stack: C++, WebRTC, GStreamer, AWS EC2, Auto Scaling Groups

Android Cloud Gaming Runtime / Streamer SDK / Controller / AOSP Customization

Built Android runtime and platform components for cloud gaming.

  • Implemented an Android cloud gaming architecture using WebRTC.
  • Developed an Android Streamer SDK and a system-signed controller app.
  • Customized AOSP for input injection from stream and worked on streaming and recording-related components.

Tech stack: Kotlin, Java, WebRTC, AOSP, REST APIs, WebSocket API, STUN/TURN

Android Cloud Gaming Storefront / Game Library / Subscription Flows

Developed and optimized the native Android client and storefront experience for a cloud gaming service.

  • Re-architected game store UI and subscription flows.
  • Improved navigation, engagement, and reduced unnecessary client-side data fetching.
  • Mentored a junior developer during client-side architecture updates.
  • Improved startup and load time by about 20%.

Tech stack: Kotlin, Java, Android UI/UX, client architecture

INAGRAL Cloud Gaming Store - Prototype

Android Virtual Gamepad

Built an Android virtual gamepad for cloud gaming use cases.

  • Supported both full-screen and overlay modes.
  • Implemented customizable layouts, drag-and-drop controls, layout presets, and Bluetooth gamepad synchronization.

Tech stack: Kotlin, Java, Android, Bluetooth, custom views, input handling

INAGRAL overlay Android gamepad INAGRAL fullscreen Android gamepad

Android DRM & RASP Security Wrapper / APK Instrumentation Pipeline

Built a wrapper and automation pipeline that modifies compiled Android APKs without source-code access.

  • Injected DRM, subscription validation, analytics, monetization, and security-related behavior into target apps.
  • Automated manifest merging, resource merging, and resource ID repair across decompiled APKs.
  • Implemented dynamic code loading and wrapper-side protections for Android applications.
  • Added a lateral overlay and side-menu system to surface native and HTML-based apps inside wrapped applications.

Tech stack: Python, Kotlin, Java, Smali, apktool, DexClassLoader, Android resource and manifest tooling

Android DRM Wrapper

Financial LLM Market Analysis Tool

Built financial news and sentiment analysis pipelines around LLM workflows.

  • Developed multiple news data pipelines, preprocessing flows, and prompt generation workflows.
  • Fine-tuned and continuously adapted LLM-based workflows to extract market trends and sentiment for specific asset classes.

Tech stack: Python, Llama, BERT, Prompt Engineering, Large Language Models (LLM), Sentiment Analysis

Blockchain Data Analysis Tool

Built an on-chain analysis tool focused on Ethereum market anomalies.

  • Used blockchain transaction data and APIs to identify early market anomalies.

Tech stack: Python, QuickNode, Ethereum, Smart Contracts, Solidity

Freelance

Full-stack Developer Remote · Jun 2019 - Aug 2021

Part-time freelance work during my studies across local clients and hobby projects. Some client work is not listed publicly due to NDA.

Representative public work: - School management system web app - Grocery shopping mobile app - Fresh fruit shopping and delivery Android app - Puzzle game Android app - Open-source Telegram Android app customization

Tech stack: Android, Python backend, PHP / Laravel, Java, Kotlin, PostgreSQL, MySQL

SedaBook

Android Developer Self-employed · Sep 2019 - Nov 2019

Worked on SedaBook, an online audiobook marketplace. Earned 2nd place at Sharif University Mobile Development Competition 2019.

Baghalimoon

Cross-Platform Mobile Developer Freelance · Mashhad, Iran · Sep 2017 - Oct 2018

Built a cross-platform React Native app for an online grocery store on Android and iOS.

  • Implemented authentication, catalog, cart, checkout, and payment flows.
  • Reached a couple thousand installs across Android and iOS, with active users in the hundreds.

Tech stack: React Native, mobile commerce

Baghalimoon

Mashhad Wheel Manufacturing Co.

Software Engineer Intern Mashhad, Iran · Jul 2018 - Sep 2018

Built an intranet communication chat application proof of concept.

  • Developed an Android client integrated with .NET APIs.
  • Delivered an MVP for internal communication workflows.

Selected Open-Source & Public Work

JobFinder

Job finder and CV creator built as a serverless application. Stack: SerpApi, Google Gemini, AWS SAM, EventBridge, SQS, DynamoDB, SES, Python Repository

LLM-Tools

Public repository for building tools for LLMs with LangGraph and Pydantic. Repository

LLM-Extension

Python project for extending an LLM with a custom functionality layer using LLaMA 3 and Ollama. Repository

tts-engine

Tacotron 2 experiments for Persian text-to-speech. Stack: PyTorch, Mozilla Common Voice Repository

fetch-awesome

React Native fetch wrapper with timeout handling, retries, request cancelation, simple cache support, and progress reporting. Repository

Writing & Experiments

Selected topics published across my website and public technical work:

  • Building a Local AI Audiobook Pipeline with Python and Kokoro
  • Building a Todo App with GCP: Part 1 — Setting Up Infrastructure
  • AI Assistant Using Llama 3 and Ollama via Python
  • Easy Subtitle Translation with Google API
  • REST API vs GraphQL on AWS — API Gateway vs AppSync
  • Streamlining API Integration with Kiota and OpenAPI
  • JAX vs NumPy performance comparison
  • How Android Implements Jetpack Compose
  • Overview of Coroutines

Core Skills

  • Languages: Kotlin, Java, Python, C++, JavaScript, SQL, PHP
  • Cloud & Backend: AWS Lambda, ECS/Fargate, SQS, DynamoDB, EventBridge, SAM, FastAPI, Node.js, REST APIs, GraphQL / AppSync, Terraform, Firestore, SES
  • Android & Mobile: Android SDK, Jetpack Compose, React Native, AOSP customization, system-signed Android apps, APK instrumentation, Smali, apktool, DexClassLoader, Bluetooth controller support
  • Real-time Systems: WebRTC, GStreamer, STUN/TURN, signaling services, session orchestration, event-driven architecture
  • AI / Data: LLM workflows, LangGraph, Pydantic, Llama / LLaMA 3, Ollama, Google Gemini, Prompt Engineering, BERT, Sentiment Analysis, PyTorch, JAX, NumPy, LDA, CNN, Random Forest
  • Datastores: DynamoDB, PostgreSQL, MySQL, Firestore, Ethereum / QuickNode data sources

Licenses & Certifications

Google Cloud — Generative AI Leader

Google Cloud Issued Sep 2025 Credential ID: KMWKKS4FJY19


Associate Cloud Engineer Certification

Google Issued Sep 2024 - Expires Sep 2027 Show credential


AWS Certified Data Engineer – Associate

Amazon Web Services (AWS) Issued Jun 2024 - Expires Jun 2027 Show credential


AWS Certified Solutions Architect – Associate

Amazon Web Services (AWS) Issued May 2024 - Expires May 2027 Show credential


Machine Learning with Python

Coursera Issued Aug 2022 Credential ID: EJNEU5KH7FD8 Show credential


Exploratory Data Analysis for Machine Learning

Coursera Issued Jun 2022 Credential ID: AMYAHE2UVL8X Show credential

Education

Bishop's University

Master of Science (MSc), Computer Science Sep 2021 - Jan 2025

Thesis: Time-Series Forecasting of ECG and EOG Signals in EEG Recordings: A Multivariate Time-Series Approach

  • Developed a multivariate regression approach to remove ECG and EOG artifacts from EEG recordings without additional sensors.
  • Achieved performance comparable to conventional methods, with PSNR around 39 dB on cleaned EEG signals.
  • Validated the approach on multiple datasets as a cost-effective solution for EEG noise filtering in clinical and research applications.

Ferdowsi University of Mashhad

Bachelor of Engineering, Computer Engineering

Undergraduate Thesis: Persian Text-to-Speech Engine

Developed a real-time Persian text-to-speech engine using Tacotron 2 and Grapheme-to-Phoneme (G2P) models for accurate pronunciation prediction.

Research Project: Detection of Autism in Twitter

Conducted research on autism detection through Twitter feed analysis using LDA topic modeling, Convolutional Neural Networks (CNN), and Random Forest classification.