Agentic.swiss / Zurich, Switzerland

Leverage compounds.
Judgment stays human.

We build AI-first systems so agents carry the operational bulk — research, content, ops loops — and people spend attention where it actually moves the company: strategy, trust, and hard calls.

See the model Read insights

RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE // RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE // RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE // RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE // RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE // RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE // RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE // RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE // RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE // RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE // RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE // RESEARCH // BUILD // REVIEW // SHIP // COMPOUND // JUDGE //

01 / The model

Agents by default on execution.
Humans by default on judgment.

Most companies run AI like an app on top of existing infrastructure. New features, same bottlenecks underneath. We didn't install an app. We redesigned the operating layer around leverage.

Agents take the grind first — the loops, drafts, monitors, and handoffs that eat operator time. Then you decide where human attention is irreplaceable: direction, relationships, taste, and final calls. Not the other way around.

As models improve, more execution moves into the stack. Judgment does not get outsourced by default. The point is a sharper power curve for a small team — compound output without pretending people are optional.

Exec

Agents by default

Judge

Humans by default

Leverage, not headcount

AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE // AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE // AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE // AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE // AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE // AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE // AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE // AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE // AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE // AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE // AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE // AGENTS ON EXECUTION // HUMANS ON JUDGMENT // MEMORY // LOOPS // LEVERAGE //

02 / How it works

The stack does the grind.
You keep the aim.

Agents handle research, content, analytics, and iteration loops end to end. People set priorities, standards, and the decisions that carry risk.

Institutional Memory

Context compounds.

Decisions, threads, and data feed a living model. Less re-briefing. More continuity across people and agents.

Execution Layer

Work moves without waiting.

Background jobs, monitors, drafts, follow-ups. Agents clear the queue so judgment is spent on what matters.

Thin Overhead

Signal over ceremony.

Agents coordinate at machine speed. Humans stay in the loop through tight reviews — not meetings for their own sake.

Feedback Loops

Rated work gets better.

Approve, edit, reject — then feed the signal back. The stack improves from real use, not slide decks.

Model Freedom

Best brain for the job.

Frontier APIs or open weights. Swap the model, keep memory and process. No single-vendor religion.

Multi-Agent

Specialists, one desk.

Research, ops, code, comms — separate agents with shared state. Orchestrated, not a single chat blob.

OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE // OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE // OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE // OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE // OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE // OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE // OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE // OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE // OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE // OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE // OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE // OBSERVE // DRAFT // DECIDE // EXECUTE // MEASURE // IMPROVE //

03 / Insights

What we're watching.

Substance briefs — mechanisms, numbers, second-order effects. Built to be useful to operators and capital, not a hype feed.

ModelsAug 14, 2026

Qwen3.8-27B: the open dump that actually runs

Alibaba kept the Max-class promise in two pieces: a 2.4T text-only checkpoint under a custom licence (Aug 12), then Qwen3.8-27B — dense, multimodal, Apache 2.0 — on Aug 14. The 27B is the operator product; Max-class 'open' is still a datacenter hobby.

SecurityAug 14, 2026

GLM-5.3: post-training produced exploit chains Z.ai didn’t plan

Same ~743B base as 5.2; every gain is scaled RL. Coding jumped; cyber jumped faster — CyberGym 84.5%, chain-level offense, 2,436 vulns across 269 OSS projects. Weights held two weeks for hardening. Open-weight labs now have a Mythos problem.

ModelsAug 13, 2026

Gemini 3.7 Flash: Google’s best model is the cheap one

Three weeks after 3.6 Flash, Google shipped 3.7 Flash at $0.75/$3.75 intro — half the prior Flash rate — and still no 3.5 Pro. DeepSWE 65.3%, first-pass coding up, Spark gets the brain. The workhorse is the flagship by default.

ModelsAug 12, 2026

Grok 4.6: post-training as the new scale-up

SpaceXAI kept the Grok 4.5 base and bought five Intelligence Index points with longer supplemental training, regenerated SFT, and agentic RL. Same $2/$6, 500K context, live in Cursor. Frontier is now a recipe, not a new pretrain.

AgentsAug 11, 2026

Grok Bot: persistent agents with their own computer

SpaceXAI and Cursor shipped Grok Bot into early beta: always-on teammates on cloud VMs that sign into your apps, keep working after you close the laptop, and only ping for approval. The product category moved from 'answer' to 'finished work in the tool.'

View all insights →

Leverage compounds.Judgment stays human.

Agents by default on execution.Humans by default on judgment.

The stack does the grind.You keep the aim.