Building Low-Latency AI Infra & Stateful Multi-Agent Backends

Specializing in Python, TypeScript/Node.js, and Rust. Architecting stateful agentic systems, async FastAPI & Node backends, sub-150ms edge inferencing, and pgvector knowledge retrieval.

Get In Touch View My Work

Tech Stack

My expertise spans across

Generative AI

Machine Learning

Edge AI

Backend & Ops

Continuously exploring new technologies and frameworks to build better solutions.

About Me

I bridge the gap between academic research (IIT Mandi) and production systems engineering. With a unique background spanning economics, public policy, and artificial intelligence, I am passionate about privacy-first local computing, minimalist Linux workflows, and building software that feels seamless and invisible.

My Mission

To build high-performance AI systems that eliminate cloud vendor lock-in, deliver sub-150ms real-time interactions, and bring privacy-first agentic intelligence directly to local environments.

Stateful Agentic Orchestration

Building resilient multi-agent graphs with LangGraph, async PostgreSQL state management, and long-term memory.

Sub-150ms Edge & Speech Pipelines

Optimizing local voice activity detection (VAD), speech inference, Rust/WASM, and ONNX Runtime execution.

Hybrid RAG & Knowledge Retrieval

Engineering production-grade RAG architectures combining vector embeddings (pgvector), keyword search, and reranking.

Decoupled Systems & Production Backends

Developing Unix domain socket daemons, low-level Linux authentication modules, and async FastAPI & Node.js backends.

Beyond the Code

When I'm not engineering AI systems, I'm usually:

Reading about public policy, economics, and how technology reshapes institutions.
Tinkering with Linux ricing, DWM configs, and minimalist developer workflows.
Exploring Japanese language and culture, which shaped how I think about precision.
Writing documentation and technical notes that actually help people.

"The best AI systems are fast, local, and invisible: users forget they are even there."

Featured Projects

Loading projects from GitHub...

Resume & Experience

My professional journey, education, and certifications in data science and AI.

Download PDF Resume

AI Systems Engineer / Developer

Independent Research & Development

2024 - PresentRemote

•Architecting production-grade agentic platforms (Disha, CodexEngine) using LangGraph, pgvector, and async PostgreSQL.

•Engineering sub-150ms speech pipelines (vad_processor) and edge Linux biometric daemons (Aura).

•Structuring reinforcement learning environments for recommendation and personalized simulation.

LangGraphFastAPIpgvectorPyTorchRust/WASMONNXTauriPython

Technical Support Executive

Teleperformance

Jul 2023 - Apr 2024Jaipur, India

•Diagnosed software issues in production environments while maintaining a CSAT of 4.8/5.

•Translated user issues into technical requirements and used SQL-based analysis to improve tool reliability and system performance.

Customer SupportSQLSalesforceJapanese Language

Freelance Developer & Technologist

Self-Employed

2022 - 2024Remote

•Delivered modular CLI utilities and custom automation scripts in Python.

•Configured and optimized minimalist, Linux-first local development workflows.

PythonCLI ToolsNeovimLinux

Get In Touch

Ready to start a conversation? Let's discuss your next project or collaboration opportunity. Or, if you'd prefer, send me a message using the form below.

Let's Connect

I'm always interested in hearing about new opportunities and exciting projects. Whether you have a question or just want to say hi, I'll try my best to get back to you!

anmolsharma152.dev@gmail.com

GitHub

github.com/anmolsharma152

linkedin.com/in/anmolsharma152