Skip to content
View ashish-doing's full-sized avatar

Highlights

  • Pro

Block or report ashish-doing

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 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
ashish-doing/README.md

Typing SVG

LinkedIn HuggingFace Email Resume Kaggle Credly


ECE undergrad at IIIT Guwahati (Batch 2028), building in AI/ML and intelligent agent systems.

  • 🎯 Focus: Reinforcement Learning · LLM Agents · MLOps · Physical AI · Robotics
  • 🛠️ Stack: Python · PyTorch · LangGraph · FastAPI · HuggingFace · OpenEnv · ROS 2
  • 📜 Certifications: Microsoft AI Skills Fest 2026 · Google × Kaggle AI Agents Intensive · IBM SkillsBuild AI Builders · e-Yantra IIT Bombay
  • 📚 Currently learning: Andrew Ng ML Specialization · Reinforcement Learning for Robotics
  • 🤖 Building: Autonomous UGV agents — Cyberwave Builders Program
  • 💡 Open to internships and collaborations

🚀 Projects

Machine Learning & AI

Title Description Technologies
openenv-supply-chain Full RL environment for supply chain optimization — 10 task variants, 118/118 validator checks · Grand Finale Meta × PyTorch × Scaler Hackathon
Live Demo
Python Docker HuggingFace
pitwall-f1-copilot F1 race strategy copilot — Q-Learning agent trained on 23,400 real lap decisions from FastF1 telemetry + IBM Granite 4.1B · IBM SkillsBuild Challenge
Live Demo
Python HuggingFace Gradio
lexscout 5-agent LangGraph legal intelligence pipeline — ingestion, chunking, embedding, retrieval, synthesis on typed state graph · Bright Data Hackathon
Live Demo
Python LangGraph FastAPI
repoterrain GitLab repo → semantic embeddings → UMAP 3D reduction → navigable terrain in Three.js · Google Cloud Rapid Agent Hackathon
Live Demo
Python FastAPI JavaScript
agentwatch Real-time MLOps monitoring for multi-agent LLM systems — OpenTelemetry instrumentation + Splunk dashboards · Splunk Agentic Ops Hackathon
Live Demo
Python Docker OpenTelemetry

Other Builds

Title Description Technologies
autostream-agent Production conversational AI agent — LangGraph state graph, RAG pipeline, multi-turn lead capture
Python LangGraph Groq
mlops-signal-pipeline MLOps batch pipeline: 10k-row OHLCV → rolling-mean signal → structured JSON metrics · Dashboard
Python Docker NumPy

🏆 Hackathons & Competitions

Result Event Project
🥇 Grand Finale Meta × PyTorch × Scaler AI Hackathon Supply Chain RL Environment
✅ Submitted IBM SkillsBuild AI Builders Challenge PitWall F1 Copilot
✅ Submitted Google Cloud Rapid Agent Hackathon RepoTerrain
✅ Submitted Bright Data Web Data Hackathon LexScout
✅ Submitted Splunk Agentic Ops Hackathon AgentWatch

📊 GitHub Stats

GitHub Stats

Top Languages

GitHub Streak


📜 Certifications

Certification Issuer Status
AI Skills Fest 2026 Microsoft ✅ Completed Jun 2026
ML Specialization Andrew Ng / Coursera 📚 In Progress
5-Day AI Agents Intensive Google × Kaggle ✅ Completed
AI Builders Challenge IBM SkillsBuild ✅ Active Member
Lab Sprint 2026 e-Yantra, IIT Bombay 📚 In Progress
Google Cloud Arcade Google Cloud 🕹️ Active Participant
Builders Program Cohort 2 Cyberwave ✅ Core Builder

📍 Guwahati, Assam | INDIA

Pinned Loading

  1. openenv-supply-chain openenv-supply-chain Public

    OpenEnv-compatible supply chain disruption environment for LLM agent training - Scaler × Meta PyTorch Hackathon

    Python

  2. agentwatch agentwatch Public

    Real-time AI agent observability for Splunk — OpenTelemetry SDK, 5-framework support, two-stage anomaly detection, Foundation-Sec explanations, and Three.js live brain visualization

    Python

  3. lexscout lexscout Public

    5-agent AI pipeline: describe any legal situation → get applicable laws, case precedents, complaint letter, filing portal & legal aid in <30s. Powered by Bright Data + Groq + LangGraph.

    HTML 1

  4. pitwall-f1-copilot pitwall-f1-copilot Public

    AI-powered F1 race strategy copilot using IBM Granite + Docling + FastF1 telemetry + Q-Learning RL agent. IBM SkillsBuild AI Builders Challenge May 2026.

    HTML

  5. repoterrain repoterrain Public

    Transform any GitLab repo into a navigable 3D semantic terrain — Gemini 2.0 Flash agent + GitLab MCP actions + MediaPipe hand tracking

    HTML