🎓 The Research Paper Roadmap by Sagar Tamang
Congratulations on taking the first step toward publishing your first AI research paper. This is the exact toolkit I used to land 5 Best Paper awards and 20+ citations.
🛠 Step 1: The Essential Toolkit
Don't use MS Word. Real research is built on LaTeX. It handles formatting, citations, and math perfectly.
- The Editor: Overleaf (Browser-based LaTeX editor).
- My Go-To Template: ArXiv LaTeX Template. This is the gold standard for AI/ML papers. Use this to make your work look professional from day one.
🔎 Step 2: Where to Find & Filter Papers
To write a great paper, you need to read what's happening right now.
- The Archive: arXiv.org — Go here for the latest AI/ML pre-prints. It’s where research hits the world first.
- The Search Engine: Google Scholar — Use this to search for specific topics and track the "impact" of a paper.
- The "Snowball" Hack: Don't just read a paper and stop.
- Look at the References: See which papers the author cited; these are the foundations of the field.
- Look at "Cited by": On Google Scholar, click "Cited by" to see newer papers that have built upon that work. This is the fastest way to find a research gap.
🧠 Step 3: The "Secret Sauce" Strategy
Before you write a single line of code, read this document. It contains the "Motown Chilly Steps" for writing a research paper. It is, hands down, the best advice I’ve ever found.
- Read this first: The Golden Guide to Research Writing
💡 Step 4: The Mindset of a Researcher (Expert Advice)
Research isn't just about writing; it's about how you think. This video features Dr. Nishanth Chandran, a Principal Researcher at Microsoft Research India.
His insights into the field of computer science research were a huge inspiration for me when I was starting out. I highly recommend watching this to understand the level of curiosity and rigor required to succeed.
- Watch here: Dr. Nishanth Chandran - Advice About Research
📧 Step 5: The "Cold Email" Template
Use this to approach professors at your university. Pro-tip: Read one of their recent abstracts on arXiv before sending this.
Subject: Undergraduate Research Opportunity - [Your Name]
Dear Professor [Last Name],
I am [Your Name], a [Year] student in the [Your Department] department. I’ve been following your work on [Mention their specific research area] and was particularly interested in your recent paper regarding [Mention a specific detail].
I am highly interested in AI/ML and would love the opportunity to contribute to any ongoing projects or research problems in your lab. I am familiar with Python and [mention any other skill like PyTorch/Next.js].
Would you be available for a brief 10-minute meeting this week to discuss any potential openings for an undergraduate researcher?
Best regards, [Your Name] [Your LinkedIn/GitHub Link]
✅ Step 6: The Literature Review Checklist
My biggest mistake was skipping this. Don't repeat it. For every paper you read, answer these 3 questions:
- What specific problem are they trying to solve?
- What is the limitation of their current approach? (This is where your idea comes from).
- How are they measuring success (Accuracy, F1 Score, Latency)?
🚀 Final Advice
Research is about consistency over intensity. Spend 30 minutes every day reading one abstract. By the end of the month, you’ll know more than 90% of your peers.
Build. Publish. Repeat.
Follow me on Instagram @sagar_builds for more AI roadmaps.