VirtualPhotography.ai
One product photo and a few reference shots in; ten on-brand product photographs out in about a minute, checked by an automated quality gate before anyone sees them.
- 1The problem
Product imagery needed studio quality and brand consistency at volume. Single-shot AI image tools fail unpredictably, cost too much per image and need someone to keep retrying.
- 2Our contribution
Designed and built by Kumar Sahil, author of almost all of the code.
- 3What we built
A production system of 8 services (gateway, auth, billing, job orchestration, live notifications, admin, web app and a Figma plugin) around a 15-stage AI pipeline: validation, normalisation, fan-out into parallel generation jobs, a durable queue with idempotency keys, model failover on timeout, moderation, and a quality gate that scores every image before delivery, streamed to the browser as results arrive.
- 4The result
In production. Results shared on request.
Other systems we’ve built
Reply Classifier Cascade
A cost-ordered classifier for English and Dutch cold-email replies that decides which replies become CRM deals, built so a false “hot deal” needs several layers to agree.Read the case studyLLM Gateway
A provider-neutral LLM gateway: code asks for a job, the gateway routes it through a chain of models with safe failover, a deadline per call, and the cost of every call written to a ledger.Read the case studyLLM Eval Gauntlet
Invariant-checked evaluation suites for tool-calling agents that run in CI and fail the build when an agent invents a number, leaks its prompt or obeys an injection.Read the case studyFind the workflow worth automating first.
Tell us about one process that costs your team hours every week. We will tell you honestly whether AI or automation is the right fix, and what a first build would take.