Skip to content
View Shreyansh15624's full-sized avatar

Block or report Shreyansh15624

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please donโ€™t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this userโ€™s behavior. Learn more about reporting abuse.

Report abuse
Shreyansh15624/README.md

Hi, I'm Shreyansh ๐Ÿ‘‹

Backend Engineer | Applied AI Architect | Systems Automation

LinkedIn Email Boot.dev

I am a software engineer based in Hyderabad, specializing in building fault-tolerant microservices, deterministic API routing, and integrating Large Language Models into resilient pipelines. I focus on bridging the gap between raw unstructured data and strict business logic via privacy-first, local-compute architectures and asynchronous agentic workflows.

Boot.dev Thumbnail

โšก Core Engineering Stack

Backend & Infrastructure AI Orchestration & LLMOps Databases & Task Queues

๐Ÿ—๏ธ Active Engineering & Systems

A high-concurrency financial ledger API handling secure user transactions. Currently scoped as a functional MVP with a fully realized Developer Experience (DX).

  • The Architecture: Built with Python, FastAPI, and asynchronous microservices. This is currently my most structurally mature project, featuring end-to-end Docker containerization for standardized cloud deployment with the start up guide smooth streamlined DX, just follow the instructions.
  • Data Integrity: Secured concurrent financial requests by integrating Redis caching & pessimistic database locks, actively preventing double-spending vulnerabilities. Offloaded heavy processing to Celery task queues.

An autonomous coding agent featuring a TUI (Rich-Textual framework) to automate workspace modifications. Currently in active Alpha development.

  • Current State: The asynchronous ReAct execution loop is active, and the TUI is successfully integrated with local Ollama daemon services, and I am going to be debugging the command execution of the AI Models, but it will happen after the Grok API is integrated into the mix, as the performance on my local hardware is a bit too sub-par ๐Ÿ˜….
  • The Roadmap: Actively debugging the TUI functionality & working on the pre-established custom JSON schema translation layer. This mapper will dynamically route payloads between remote Gemini APIs, Grok AI, and local models to strictly prevent vendor lock-in.

A microservices-based search engine querying local markdown notes.

  • Current State: The backend logic utilizing LangChain and ChromaDB is established, but the project is currently undergoing a complete structural pivot. I am actively migrating the frontend architecture away from the Reflex framework to integrate natively with Open WebUI for a more robust user experience, because I am more of a backend dev, than a frontend dev.

๐Ÿ“ˆ Current Focus

  • Expanding expertise in Agentic Frameworks and enterprise-level vector infrastructure.
  • Grinding backend architecture concepts and system design patterns on the Boot.dev training grounds.

Pinned Loading

  1. mutex-mint mutex-mint Public

    A containerized financial backend (FastAPI, PostgreSQL, Redis, Celery) featuring strict RBAC, asynchronous queues, and row-level locking for zero-race-condition transfers.

    Python

  2. synapx-tha synapx-tha Public

    Autonomous CLI agent for deterministic FNOL data extraction and claim routing using a dual-engine architecture (Google GenAI & local Ollama).

    Python

  3. aegis-agent aegis-agent Public

    A modular terminal AI agent capable of autonomous code execution, context propagation, and secure OS sandboxing using Gemini and local Ollama engines.

    Python 1

  4. marimo_comp marimo_comp Public

    An interactive, serverless WebAssembly dashboard demonstrating the statistical fragility of AI interpretability tools. Built for the alphaXiv Hackathon to simulate the multiple comparisons problem.

    Python

  5. obs-rag obs-rag Public

    An AI-powered Second Brain that lets you chat with your Obsidian notes. Built with Python, LangChain, ChromaDB, and Google Gemini 2.5.

    Python 1

  6. ssg ssg Public

    A robust, custom-built static site generator written in Python. This engine recursively crawls a directory of Markdown content, parses the text into HTML nodes, and generates a full static website โ€ฆ

    Python