DEV Community

PRANJUL RATHOUR profile picture

PRANJUL RATHOUR

GenAI Engineer & AI Product Builder | Generative AI • LLMs • RAG • AI Agents • Fine-Tuning • QLoRA/LoRA • Multimodal AI • LLM Evaluation • Computer Vision • AI Automation | Python • FastAPI • Next.js

5 Production AI Apps, All Open Source

5 Production AI Apps, All Open Source

Comments
2 min read
15 Free Tools, No Signup, Every Link

15 Free Tools, No Signup, Every Link

Comments
2 min read
15 Free Tools, No Signup, Every Link

15 Free Tools, No Signup, Every Link

Comments
2 min read
Building your personal brand as a CS student

Building your personal brand as a CS student

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

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

Comments
2 min read
NIMCET, placements, or a startup: choosing your Year 3

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

Comments
2 min read
How to start a tech community on your campus

How to start a tech community on your campus

Comments
1 min read
Writing your first technical blog post: a template

Writing your first technical blog post: a template

Comments
1 min read
Beating 15+ teams at BYTEBATTLE with Annapurna

Beating 15+ teams at BYTEBATTLE with Annapurna

Comments
1 min read
Cold-emailing a mentor or speaker (and actually getting a yes)

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

Comments
1 min read
From mentee to mentor to judge: a natural progression, not a hierarchy

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

Comments
2 min read
Running a judges-only debrief right after the event: the fifteen minutes organisers skip

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

Comments
2 min read
Speaking about AI to non-engineering college audiences: what actually needs to change

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

Comments
2 min read
What I wish someone had told me before my first hackathon judging gig

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

Comments
2 min read
Building a personal brand as a campus speaker, without it feeling like self-promotion

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

Comments
2 min read
Judging a hackathon where you're also personally sponsoring a prize

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

Comments
2 min read
How organisers should train first-time judges before the event, not during it

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

Comments
2 min read
Mentoring through open source: a different channel from talks and workshops

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

Comments
2 min read
Speaking at international student conferences remotely: what changes across borders

Speaking at international student conferences remotely: what changes across borders

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

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

Comments
2 min read
Handling long-running AI agents: timeouts, checkpoints and partial progress

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

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

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

Comments
2 min read
Versioning prompts like you version code: a simple system that pays off fast

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

Comments
2 min read
Graceful degradation: what your app should do when the LLM provider is down

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

Comments
2 min read
Setting up cost alerts for AI APIs before the bill surprises you

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

Comments
2 min read
Webhooks and async callbacks for AI pipelines: notifying instead of polling

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

Comments
2 min read
Building a support-ticket triage system with an LLM: a project walkthrough

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

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

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

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

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

Comments
2 min read
Feature flags for AI features: rolling out a new model or prompt safely

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

1
Comments
2 min read
Generating alt text with vision AI: a genuinely useful accessibility application

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

Comments
2 min read
Video summarization with vision-language models: what's actually feasible today

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

Comments
2 min read
Building a plant or object identifier app: a well-scoped vision project

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

Comments
2 min read
Annapurna: from a 36-hour hackathon prototype to a real food-sharing platform

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

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

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

Comments
2 min read
Synthetic image data: when it's a legitimate fix for a scarce dataset

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

Comments
2 min read
Vision Transformers, explained for practitioners who just want to use one well

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

Comments
2 min read
Barcode and QR detection in noisy real-world images: what actually helps

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

Comments
2 min read
Document tampering detection: what a vision model can and can't catch

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

Comments
2 min read
Testing a vision model for bias across skin tones before you ship it

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

Comments
2 min read
Building a simple attendance system with face recognition: a project walkthrough

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

Comments
2 min read
Evaluating fine-tuning ROI: was the project actually worth the compute and time?

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

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

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

Comments
2 min read
Transfer learning vs fine-tuning: are these the same thing?

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

Comments
2 min read
Estimating fine-tuning cost before you start, so there are no surprises

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

Comments
2 min read
Continual fine-tuning: updating a model over time without starting from scratch

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

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

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

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

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

Comments
2 min read
Fine-tuning an open-weight model vs a closed API's fine-tuning offering

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

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

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

Comments
2 min read
Fine-tuning a model for a specific tone of voice: what actually transfers

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

Comments
2 min read
Duplicate and near-duplicate documents in a RAG index: the quiet quality killer

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

Comments
2 min read
RAG over a codebase: why retrieving code needs a different chunking strategy

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

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

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

Comments
2 min read
RAG chatbot vs RAG-powered search: same technique, different product

RAG chatbot vs RAG-powered search: same technique, different product

Comments
2 min read
Evaluating retrieval quality when you have no labelled query-document pairs

Evaluating retrieval quality when you have no labelled query-document pairs

Comments
2 min read
RAG over mixed Hindi-English documents: what breaks and what to check

RAG over mixed Hindi-English documents: what breaks and what to check

Comments
2 min read
Multi-tenant RAG: making sure one customer never sees another customer's documents

Multi-tenant RAG: making sure one customer never sees another customer's documents

Comments
2 min read
The RAG security hole nobody checks: prompt injection hiding inside retrieved documents

The RAG security hole nobody checks: prompt injection hiding inside retrieved documents

Comments 1
2 min read
Incremental indexing: keeping a RAG index current without re-embedding everything

Incremental indexing: keeping a RAG index current without re-embedding everything

Comments
2 min read
loading...