Skip to content
View aakash-priyadarshi's full-sized avatar
🌍
Working from home
🌍
Working from home
8 followers · 4 following

Achievements

Achievement: Pull Sharkx2Achievement: Pair Extraordinairex2Achievement: Quickdraw

Achievements

Achievement: Pull Sharkx2Achievement: Pair Extraordinairex2Achievement: Quickdraw

Block or report aakash-priyadarshi

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’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 user’s behavior. Learn more about reporting abuse.

Report abuse
aakash-priyadarshi/README.md

Aakash Priyadarshi at his Alienware workstation — Full-Stack Engineer and AI / RL researcher. Animated scene with teal and purple lighting. Motif: BUILD, VERIFY, SHIP.

Animated terminal: whoami Aakash Priyadarshi; focus AI, RL, Full Stack; build, verify, ship. Agent to verifier loop. AI has to be correct, not just confident.

Aakash Priyadarshi

Full-Stack Engineer · AI / RL Researcher


Currently building ViharaOS, Gold Shop, and agent verification research



Portfolio   Email



Profile views Status Location

About

name: Aakash Priyadarshi
role: Full-Stack Engineer · AI / RL Researcher
location: London, UK
portfolio: https://aakash-priyadarshi.github.io
focus:
  - Safety-aware AI evaluation & policy gates
  - Long-horizon RL environments for coding agents
  - Production web apps (TypeScript, Python, MERN)
currently_building:
  - ViharaOS
  - Gold Shop platform
  - Frontier agent verification research
ask_me_about: MERN, PyTorch, offline RL, FastAPI, agentic systems
reach_me: aakashm301@gmail.com

What I do

I build full-stack products and research systems where AI has to be correct, not just confident — from safety-aware evaluation platforms to long-horizon RL environments for coding agents.

Highlights

  • Google Scholar profile with published research
  • Hackathon-built safety evaluation platform (SkyGuard AI)
  • Open-source work across TypeScript, Python, and Flutter
  • 50+ public repositories since 2017

Featured work

SkyGuard AI preview

SkyGuard AI

Safety-aware training & evaluation for sequential policies & LLMs

Python PyTorch FastAPI Next.js Offline RL

Deterministic policy gates, strict verifiers, and simulation-backed evaluation for frontier models.

 ViharaOS preview

ViharaOS

Platform + desktop experience

TypeScript HTML

Web platform with a dedicated desktop client — end-to-end product engineering.
Frontier env preview

Frontier Long-Horizon Env

Paired-incident substrate for coding-agent verifier research

Python

Deterministic environments for studying long-horizon agent behavior and verification.

 Jarvis preview

Jarvis

Voice-enabled OpenAI assistant

Node.js OpenAI Speech-to-Text Text-to-Speech

Real-time voice ↔ text interface powered by OpenAI and Google Cloud speech APIs.

→ See all repositories  ·  → Full portfolio site

Tech I reach for

Frontend

TypeScript, JavaScript, React, Next.js, HTML, CSS, Tailwind CSS, Redux

Backend

Node.js, NestJS, Express, Python, FastAPI

REST APIs WebSockets gRPC

AI / ML

PyTorch, TensorFlow

OpenAI Claude Gemini Qwen LangChain RAG Vector databases

Databases / Storage

PostgreSQL, MongoDB, Redis, Prisma

Qdrant Valkey

Cloud / DevOps / Infra

AWS, GCP, Azure, Docker, GitHub Actions, Vercel, Cloudflare

Railway

Coding time

This week

From: 30 August 2026 - To: 06 September 2026

Total Time: 34 hrs 53 mins

TypeScript     13 hrs 25 mins        █████████▒░░░░░░░░░░░░░░░   37.08 %
Markdown       11 hrs 14 mins        ███████▓░░░░░░░░░░░░░░░░░   31.04 %
Python         4 hrs 5 mins          ██▓░░░░░░░░░░░░░░░░░░░░░░   11.33 %
Text           3 hrs 25 mins         ██▒░░░░░░░░░░░░░░░░░░░░░░   09.46 %
Other          1 hr 18 mins          █░░░░░░░░░░░░░░░░░░░░░░░░   03.61 %
CSS            31 mins               ▒░░░░░░░░░░░░░░░░░░░░░░░░   01.47 %
HTML           24 mins               ▒░░░░░░░░░░░░░░░░░░░░░░░░   01.11 %
JavaScript     20 mins               ▒░░░░░░░░░░░░░░░░░░░░░░░░   00.96 %
JSON           20 mins               ▒░░░░░░░░░░░░░░░░░░░░░░░░   00.95 %
Prisma         17 mins               ▒░░░░░░░░░░░░░░░░░░░░░░░░   00.79 %

All time

From: 27 May 2026 - To: 06 September 2026

Total Time: 102 hrs 48 mins

TypeScript     50 hrs 16 mins        ████████████░░░░░░░░░░░░░   47.86 %
Markdown       13 hrs 35 mins        ███▒░░░░░░░░░░░░░░░░░░░░░   12.94 %
Python         12 hrs                ███░░░░░░░░░░░░░░░░░░░░░░   11.43 %
JavaScript     5 hrs 19 mins         █▒░░░░░░░░░░░░░░░░░░░░░░░   05.08 %
YAML           4 hrs 42 mins         █░░░░░░░░░░░░░░░░░░░░░░░░   04.47 %
Text           3 hrs 59 mins         █░░░░░░░░░░░░░░░░░░░░░░░░   03.81 %
Bash           2 hrs 35 mins         ▓░░░░░░░░░░░░░░░░░░░░░░░░   02.46 %
Other          2 hrs 15 mins         ▓░░░░░░░░░░░░░░░░░░░░░░░░   02.14 %
PowerShell     2 hrs 7 mins          ▓░░░░░░░░░░░░░░░░░░░░░░░░   02.03 %
JSON           1 hr 57 mins          ▒░░░░░░░░░░░░░░░░░░░░░░░░   01.86 %

GitHub activity

streak stats github stats



top languages pinned skyguard



profile details



contribution chart

3D contribution calendar

3D Contribution Calendar

Contribution snake

github contribution grid snake animation

Connect

LinkedIn Google Scholar Kaggle HackerRank Portfolio

Let's build, verify, and ship something ambitious.


Portfolio CTA

Pinned Loading

  1. viharaos-desktop viharaos-desktop Public

    desktop version of the web app viharaos.com

    HTML

  2. phd-apply-agent phd-apply-agent Public

    Python 1 1

  3. Dark-matter-mapping-model Dark-matter-mapping-model Public

    Developing an innovative image processing pipeline that analyzes weak gravitational lensing data through deep learning.

    Python 1

  4. frontier-long-horizon-env-trial frontier-long-horizon-env-trial Public

    Deterministic paired-incident substrate for long-horizon coding-agent verifier research.

    Python

  5. gold-shop-app gold-shop-app Public

    TypeScript

  6. skyguard-ai-build-week-submission skyguard-ai-build-week-submission Public

    Safety-aware training and evaluation platform for sequential policies and frontier LLMs under a deterministic policy gate and strict verifier.

    Python