ashutosh-profile
🛕

Greetings, I am Ashutosh

an Software Engineer

24, he/him

about me.

Hi there! I am Ashutosh Jha, a tech enthusiast and a software engineer. I love building stuff that makes people's lives easier. I am currently working as a Software Engineer at Paytm. I am a quick learner and I am always looking for new challenges.

I am currently open for full-time Software engineering roles.

Are you a collaborative team of engineers or a developers with cool ideas who are building great products. Interested in working together? Feel free to schedule a meet!

worked as.

Software Engineer, Paytm

at, Paytm

January, 2025-Present

  • Build and maintain distributed backend services for Paytm Soundbox IoT payment devices, supporting real-time transaction processing across millions of merchants and 12+ device variants in India and Indonesia.
  • Design real-time, pub/sub-based data pipelines (MQTT) for RTOS-based devices, delivering sub-second event delivery at scale across a distributed device fleet handling millions of daily transactions.
  • Architected a low-latency audio streaming pipeline for AI Soundbox using phase-wise chunking, Base64-encoded PCM buffering, and a 10K-entry prefetch buffer, reducing decoding overhead versus the prior AMR/MP3 approach.
  • Designed a Root CA certificate validation and rotation system for secure device authentication across the fleet, working closely with security and infra teams.
  • Designed and built an end-to-end agentic AI development pipeline — architect, developer, reviewer, and validator agents coordinating autonomously from a single Jira ticket through code, review, and deployment — cutting feature planning-to-deployment time from 2 weeks to 4 days.

Software Developer Intern, Denr Financial Services

at, Denr

March, 2024-June, 2024

  • Built a cross-platform mobile app in React Native with authentication, profile management, and financial data visualization, taking it from POC to a production-ready application.
  • Integrated AWS Cognito, S3, and Lambda; improved backend API response times by 40% through async processing and caching.
  • Wrote a Jest-based automated test suite reaching 85% coverage and participated in code review, reducing regression bugs shipped to production.

Frontend Development Intern, House Technologies

at, House Technologies

October, 2023-December, 2023

  • Drove website development and maintenance initiatives, harnessing the power of React.js to craft dynamic and user-friendly web solutions.
  • Collaborated closely with cross-functional teams, actively gathering and analyzing project requirements to ensure alignment with client expectations.
  • Showcased proficiency in a diverse array of programming languages and frameworks, such as HTML, CSS, JavaScript, and React.js, to engineer and optimize website features. Demonstrated a proactive approach in identifying and resolving website issues, ensuring a seamless and error-free user experience.

Microsoft Cybersecurity Engage Mentee, Microsoft

at, Microsoft Engage 2022

June, 2022-Sept 2022

  • Participated in an intensive 6-week program to develop skills in cybersecurity and threat intelligence. Completed hands-on exercises, identified and remediated security threats, and collaborated on a real-world cybersecurity project. Improved skills in threat hunting, vulnerability management, and incident response.

cool projects.

SysMCP — Linux Debugging Primitives as Agent Tools

GitHub
🏗️ Ongoing
  • An MCP server exposing real Linux production-debugging primitives — perf, strace, /proc, flamegraphs and eBPF — as safe, structured, agent-callable tools.
  • Lets an agent autonomously diagnose why a service is slow instead of reading logs and guessing.
category: Dev-Tools, Systems

DriftBench — Measuring Context Drift in Coding Agents

GitHub
🏗️ Ongoing
  • Asks whether coding agents get less reliable as a codebase evolves, and whether structured memory plus verification fixes it.
  • Runs an agent through 30 sequential tickets against one growing codebase, instrumenting where and why it drifts, measured against a naive sliding-window baseline.
category: AI/ML, Dev-Tools

even cooler projects.

OktaDB — Embedded Relational Database Engine

GitHub
🏗️ Ongoing
  • Built a relational database engine from scratch in C with a custom storage engine, page-based persistence layer, and B-Tree indexing for O(log n) reads, inserts, and updates.
  • Designed a SQL-like query executor supporting INSERT, SELECT, DELETE with predicate filtering, with no external libraries.
category: Core-CS

ThrottleX — High-Throughput Rate Limiting Service

GitHub
🏗️ Ongoing
  • Built a rate-limiting microservice implementing Token Bucket and Sliding Window algorithms, exposing REST APIs designed for high-throughput, low-latency use.
  • Containerized with Docker and exposed monitoring endpoints for real-time observability of token state and throughput.
category: Core-CS

Push2Prod — Automated Cloud Deployment Platform

Live PreviewGitHub
👍🏽 Maintained
  • Built a Git-to-cloud deployment platform with Redis-backed task queues coordinating concurrent deployment workflows across EC2, EKS (Kubernetes), and S3.
  • Integrated GitHub Actions CI/CD pipelines with AWS infrastructure for automated build, test, and deploy workflows.
category: Full-Stack

QuantCraft — C++ Backtesting Engine

GitHub
👍🏽 Maintained
  • A lightweight, modular backtesting skeleton in modern C++17, wiring a data-loader → strategy → execution → portfolio pipeline.
  • Ships example mean-reversion and momentum strategies over a CSV-backed market data loader.
category: Core-CS, Systems

LD-Net Lightweight Image Dehazing Algorithm

GitHub
👍🏽 Maintained
  • Developed a deep learning-based algorithm to enhance visibility in hazy images and videos, improving visual quality and clarity.
  • Designed and implemented an autoencoder architecture using PyTorch to learn the dehazing process, achieving state-of-the-art performance.
  • Integrated the dehazing model with a web platform using React.js and connected it via a Flask API to provide a user-friendly interface for uploading and processing images.
category: AI/ML

Memory Doctor — AI Leak Detection for Embedded C

GitHub
🏗️ Ongoing
  • A static-analysis CLI for embedded C codebases such as FreeRTOS, using tree-sitter AST parsing to isolate functions dealing with memory allocation and IPC queues.
  • Tracks pointers as they escape local scope into RTOS queues, understanding the producer-consumer pattern across threads.
  • Feeds only targeted code blocks to an LLM to catch early returns, unhandled timeouts and conditional leaks that traditional static analyzers miss, keeping the context window small.
category: Dev-Tools, AI/ML

contact, socials.

GitHub@thealonemusk
Twitter@thealonemusk
LinkedInlinkedin.com/in/thealonemusk
Instagram@thealonemusk
Schedule a meetingcal.com/thealonemusk
Resumethealonemusk.vercel.app/resume
And I don't want the world to see me cause I don't think that they'd understand
Made of stardust, dreams, and a whole lot of love

Ashutosh