Quelm
Work with us
πŸ‡¬πŸ‡§UK-based AI studio
AI Software Studio

This is the future of how software gets built.

Quelm builds AI-powered software β€” for our clients and for ourselves.We don't consult. We ship.

agent.ts
import { Agent } from 'quelm'
const agent = new Agent({
model: 'gemma4:e2b',
offline: true
})
await agent.run()
// βœ“ complete
Live signals
streaming
Burnout risk detected9 / 10
API calls blocked2,847
Cost preventedΒ£1,240
Entries analysed14
Model layer Β· Gemma 4
POST
/v1/analyse
GET/health2ms
POST/v1/score18ms
GET/v1/trend5ms
DEL/v1/entry3ms
⚑
Budget limit reached
Agent 'scraper-v2' blocked
limit: Β£10.00 / day
spent: Β£10.00 ↑
calls: BLOCKED βœ•
TokenCap enforced in 47ms
Local storage
journal.json
offline onlyβœ“ private
TokenCap
LIVE
220+ models Β· sub-50ms enforcement
Daily budgetΒ£25.00
Calls blocked today2,847
Cost preventedΒ£1,240
Buildpassed12s
Test suitepassed34s
Deploy to prodrunning...
Production Β· v1.0.0 Β· ↑ 100% uptime
Voice Β· Whisper
πŸŽ™
Offline Β· 72MB Β· local
99 languages
Burnout risk Β· 9/10 Β· High
β€œexhausted” mentioned 4Γ— this week β€” up from 1Γ—.
✦ Step outside for 5 minutes today.
Pattern confidence: High Β· 9 signals
Quelmassembling...
gathering components...

We build software that
didn't exist yesterday.

Custom AI software

We design and build bespoke AI tools for companies who need something that doesn't exist yet. No templates. No off-the-shelf. Just the exact tool your team needs, built properly.

Our own products

We ship software under the Quelm name. If we see a problem worth solving, we build the solution. We eat our own cooking β€” everything we make for clients, we've used ourselves.

AI strategy

We help teams understand what's possible, what's worth building, and what to avoid. Straight answers, no jargon, rooted in what we've actually shipped.

Shipped under our name.
Used in the real world.

TokenCap

AI cost control. Actually enforced.

Live

TokenCap prevents AI cost explosions by enforcing hard spending limits on API calls β€” blocking them before they happen, not alerting after the money is gone. Works with 220+ models across all major providers. Sub-50ms enforcement.

πŸ”ŒOpenAI, Anthropic, Gemini + 220 more
⚑Sub-50ms enforcement
✦Free plan available
πŸ’·From Β£25/month

Quietburn

Burnout detected. Privately.

Open source

An offline AI journaling companion that watches your entries over time and warns you about burnout before you notice it yourself. Runs entirely on your device β€” no data ever leaves your machine.

πŸ”’Fully offline
πŸ€–Powered by Gemma 4 via Ollama
πŸŽ™Voice journaling
🌍Multilingual

Our principles.
Non-negotiable.

01

Responsible AI

We use AI where it genuinely helps β€” not because it's fashionable. Every system we build is explainable, auditable, and does what it claims to do. No black boxes, no hype.

02

Explainable AI

Every output references its evidence. No black boxes, no "the AI said so". Every insight is traceable back to the data that generated it.

03

Ships fast

We make decisions quickly and build the minimum that works brilliantly. We'd rather put something real in your hands than perfect something nobody's used.

04

Built to last

SaaS products are engineered for reliability and scale. Custom work is yours to own β€” no lock-in, no dependencies on us. Either way, you get something that works long after we're done.

Work with us β€”
or be first to know.

Whether you need custom AI software or want early access to what we're building, we'd like to hear from you.

Custom projects

Build with Quelm

Tell us what you're building. We'll be honest about whether we can help.

Early access

Product waitlist

Get early access to TokenCap and Quietburn. We keep the list small and go deep with early users.

Available products

TokenCap

AI spend control

Live

Quietburn

Offline burnout detection

Open source