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

Pranjul Rathour
Pranjul Rathour

Originally published at https://pranjulrathour.scult.in/blog/feature-flags-for-ai-features-safe-rollouts. That copy is the canonical version and gets updates first.

Shipping a new prompt version or a new underlying model to every user at once means a regression — a quality drop, a broken output format — reaches everyone simultaneously, with no easy way back except another deploy.

A safer rollout pattern

  • Gate the new version behind a feature flag, enabled for a small percentage of traffic first.
  • Compare quality metrics between the old and new version on real traffic before widening the rollout.
  • Keep the flag toggle-able without a redeploy, so a bad version can be turned off in seconds, not after a fix-and-redeploy cycle.

Why this matters more for AI features than typical features

A traditional bug is usually deterministic and obvious; an AI quality regression can be subtle — slightly worse answers, not crashes — and easy to miss until it's affected a lot of users. Gradual rollout with real comparison catches this before it's a wide-scale problem.

See A/B testing an AI feature.

About Pranjul Rathour

Pranjul Rathour on stage presenting a requirements-gathering and user-flow slide Requirements gathering, on stage

Presenting Annapurna on stage
Presenting Annapurna on stage

Pranjul Rathour seated in a black jacket and white turtleneck with an event lanyard Pranjul Rathour

Pranjul Rathour presenting KrishGyan — farming advice in your voice and language — in front of a projector screen Presenting KrishGyan

Pranjul Rathour, GenAI engineer from Kanpur, in a white turtleneck and black jacket, looking to the side Pranjul Rathour — GenAI engineer, Kanpur

Pranjul Rathour speaking into a microphone on stage at a MeetKats event Speaking at a MeetKats event

Walking a room through the products he has shipped
Walking a room through the products he has shipped

Pranjul Rathour is a GenAI engineer from Kanpur, India, and CTO at SCULT INDIA, currently shipping production RAG, fine-tuning and agentic AI systems, mentoring 200+ students through TechVerse Enclave, and judging and speaking at student hackathons across India. Updated 2026-09-11.

Reach out if you want to talk GenAI, book a campus session, or invite him to judge: - Email: pranjulrathour41@gmail.com - Invite / talk menu: https://pranjulrathour.scult.in/invite - Portfolio & blog: https://pranjulrathour.scult.in - LinkedIn: https://www.linkedin.com/in/pranjul-rathour/ - X: https://x.com/PranjulRathourx - Instagram: https://www.instagram.com/pranjulrathour.in/ - Bluesky: https://bsky.app/profile/pranjulrathour.bsky.social - GitHub: https://github.com/Pranjulrathour

Pranjul Rathour · GenAI engineer, 3x hackathon winner, campus mentor. Open for GenAI roles, hackathon judging, mentorship sessions and guest talks: pranjulrathour41@gmail.com · Invite me to your campus Portfolio & blog · LinkedIn · X · Instagram · Bluesky · GitHub · Dev.to

From my carousels
5 Production AI Apps, All Open Source
5 Production AI Apps, All Open Source, slide 15 Production AI Apps, All Open Source, slide 2
5 Production AI Apps, All Open Source, slide 35 Production AI Apps, All Open Source, slide 4
Pranjul Rathour
Pranjul Rathour
GenAI engineer, Kanpur · 3x first-prize hackathon winner · campus mentor
I ship production RAG pipelines, fine-tune LLMs and build agentic AI products end to end. I lead engineering at SCULT INDIA for a 14-member team and have mentored 200+ students through TechVerse Enclave.
Open to: GenAI roles, hackathon judging, mentorship sessions and guest talks at colleges.
On stage, at hackathons and on campus
Pitching at CSJMU, Kanpur
Pitching at CSJMU, Kanpur
Presenting BrandHive
Presenting BrandHive
Presenting to a room
Presenting to a room

Comments

Post a Comment

Popular posts from this blog

Forming a hackathon team: roles, skills and the mistake most teams make

Image
At an Integral Startup Foundation hackathon Originally published at https://pranjulrathour.scult.in/blog/hackathon-team-roles-and-formation . That copy is the canonical version and gets updates first. Team TechVerse — Raman Shukla, Saksham Gupta and me — won three first prizes in 2025 not because we were the strongest coders in the room, but because we had stopped overlapping. When I mentor teams before a hackathon, formation is the first thing I fix. In a packed college auditorium The four roles Builder — owns the core feature and the demo path. Usually backend or the AI piece. Interface — owns what the judge sees: the front end, the flow, the seeded data. Storyteller — owns the problem statement, the pitch, the slides and the Q&A plan; also does user research in hour one. Integrator — keeps the pieces working together, manages the repo, runs the demo rehearsals. In a team of three, the builder or interface person takes this. The mistake Five friends who all write R...
Read more

Hello from Kanpur: what I build, and what I'll write about here

Image
On the mic Originally published at https://dev.to/pranjulrathour/hello-from-kanpur-what-i-build-and-what-ill-write-about-here-f2j . That copy is the canonical version and gets updates first. I'm Pranjul Rathour, a GenAI engineer from Kanpur, India. This is the first post on this blog, so here is who is writing, what I have actually shipped, and what you can expect if you follow along. What I build Between February and July 2026, at Next Upgrad Web Solutions, I designed, built and shipped five production AI applications end to end, alone, and deployed all of them with Docker, CI and observability: RAG.NextUpgrad : a retrieval-augmented generation service with hybrid dense plus BM25 retrieval, cross-encoder reranking, streaming cited answers, and a confidence gate that refuses to call the model when the evidence is weak. It runs in about 220 MB of memory, inside a free 512 MB instance. FineTune Studio : fine-tune open models with QLoRA in a browser. Upload a dataset, validate ...
Read more

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

Image
At an Integral Startup Foundation hackathon Originally published at https://pranjulrathour.scult.in/blog/reward-modeling-basics-for-student-projects . That copy is the canonical version and gets updates first. A reward model is trained to score how good a response is, so it can guide further training of another model — the mechanism behind RLHF-style alignment. It's genuinely complex to build well, and most student projects don't need to build one. What it actually requires A dataset of paired responses with human or automated preference judgements — which output is better, and why. A separate training run to fit a model that predicts that preference, before it's ever used to guide anything else. Careful evaluation of the reward model itself, since a flawed reward model teaches the wrong thing to whatever it later guides. When a student project actually needs this Almost never, directly. Supervised fine-tuning or DPO on direct preference pairs gets most projects m...
Read more