The blog · 286 articles

Notes from the build.

Build logs from real systems, hackathon stories from both sides of the judging table, and guides for students moving into AI engineering. 286 articles across 8 topics.

Notes from the build.

From mentee to mentor to judge: a natural progression, not a hierarchy

Running a judges-only debrief right after the event: the fifteen minutes organisers skip

Speaking about AI to non-engineering college audiences: what actually needs to change

What I wish someone had told me before my first hackathon judging gig

Building a personal brand as a campus speaker, without it feeling like self-promotion

Judging a hackathon where you're also personally sponsoring a prize

How organisers should train first-time judges before the event, not during it

Mentoring through open source: a different channel from talks and workshops

Why my RAG platform says "I don't know" — building RAG.NextUpgrad

Speaking at international student conferences remotely: what changes across borders

Hackathon sponsors and judging: staying neutral when a sponsor's tech is on the table

Fine-tuning a 1.7B model at 3.2 GB VRAM — building FineTune Studio

A year of shipping: SCULT INDIA, TechVerse Enclave, and NextUpgrad

Handling long-running AI agents: timeouts, checkpoints and partial progress

Face recognition that never uploads a face — building FaceVision

Building an internal AI tool for your own team: a lower-stakes way to ship something real

Versioning prompts like you version code: a simple system that pays off fast

Reading an invoice like an auditor — building DocuLens AI

Graceful degradation: what your app should do when the LLM provider is down

Setting up cost alerts for AI APIs before the bill surprises you

Every page and every word, searchable — building the OCR & Speech Workspace

Webhooks and async callbacks for AI pipelines: notifying instead of polling

Building a support-ticket triage system with an LLM: a project walkthrough

Shipping side projects that survive demo day

Handling personal data (PII) in LLM applications without it leaking where it shouldn't

A/B testing an AI feature: what to actually measure beyond click-through rate

Agentic AI patterns: tools, memory and guardrails without the hype

Feature flags for AI features: rolling out a new model or prompt safely

Secrets management for student projects: never commit an API key again

Generating alt text with vision AI: a genuinely useful accessibility application

Video summarization with vision-language models: what's actually feasible today

Controlling LLM costs: token budgets, caching and cheaper routing

Building a plant or object identifier app: a well-scoped vision project

Building SCULT Studio, the internal CRM our own agency runs on

Vision API costs: cloud vision APIs vs self-hosting a model, a real comparison

Testing LLM applications: unit tests, golden sets and regression prompts

Annapurna: from a 36-hour hackathon prototype to a real food-sharing platform

Synthetic image data: when it's a legitimate fix for a scarce dataset

Vision Transformers, explained for practitioners who just want to use one well

FastAPI + Next.js: the architecture I teach for every AI product

Cold-emailing a mentor or speaker (and actually getting a yes)

Beating 15+ teams at BYTEBATTLE with Annapurna

Barcode and QR detection in noisy real-world images: what actually helps

Document tampering detection: what a vision model can and can't catch

What I'd tell my first-year self about AI, college and shipping

Writing your first technical blog post: a template

Testing a vision model for bias across skin tones before you ship it

Building a simple attendance system with face recognition: a project walkthrough

Why I write: notes, blogs and building an audience that invites you

How to start a tech community on your campus

Evaluating fine-tuning ROI: was the project actually worth the compute and time?

Fine-tuning a vision-language model: what's different from fine-tuning a text-only LLM

The first 500 members: how TechVerse Enclave grew

NIMCET, placements, or a startup: choosing your Year 3

Three times I didn't win first place (and what stuck more than the wins)

Transfer learning vs fine-tuning: are these the same thing?

Leading a 14-member team while still a student

Building your personal brand as a CS student

Estimating fine-tuning cost before you start, so there are no surprises

Continual fine-tuning: updating a model over time without starting from scratch

What three hackathon wins (and several losses) taught me about building

How to pitch a startup idea in 2 minutes (and why most pitches fail)

Fine-tuning for reliable structured output (JSON mode isn't always enough)

Deduplicating a fine-tuning dataset before training (and why near-duplicates matter more than exact ones)

Mistakes from my first year of shipping AI products

Portfolio over certificates: what recruiters actually look at

Pitching BrandHive to a Vice Chancellor — winning Product Genesis at CSJMU

Fine-tuning an open-weight model vs a closed API's fine-tuning offering

Reward modeling, explained simply: what it is and why most student projects don't need it

From engineering student to GenAI engineer at an AI studio: what actually happened

How to win your first hackathon: a practical playbook

Fine-tuning a model for a specific tone of voice: what actually transfers

Duplicate and near-duplicate documents in a RAG index: the quiet quality killer

Why I became a GenAI engineer instead of 'just' a web developer

AI tools every CS student should learn in 2026

RAG over a codebase: why retrieving code needs a different chunking strategy

How much retrieved text should actually go in the prompt? A context-budget framework

Building a tech career from a tier-2 Indian city: communities, events and what I'd do again

From notebook to production: a checklist for shipping AI apps