The Causal Knowledge Company crafting AI products that teams can trust from sprint one.
We fuse LLM mastery, agentic development, knowledge graphs, and causal reasoning to build production-grade AI applications in weeks. Bring us the problem; we deliver the MVP, the ops stack, and the compliance story.
Proven build patterns
Reference architectures for agentic apps, decision intelligence dashboards, and knowledge copilots. All pre-wired with observability, governance, and multi-modal UX.
Full-stack causal AI delivery at startup velocity
Deep technical craft across the entire stack—product strategy, design, engineering, and AI research—so your MVP feels inevitable.
LLM & multimodal intelligence
Strategic model selection, fine-tuning, and guardrailing so AI outcomes stay accurate, safe, and on-brand.
- Custom instruction-tuned models tailored to domain language
- Evaluation harnesses covering hallucination, bias, and drift
- Realtime moderation and retrieval-in-the-loop safeguards
Agentic product engineering
Composable agent swarms orchestrated around goals, with deterministic fallbacks where reliability matters most.
- Planner-executor stacks with human-in-the-loop review
- Simulation sandboxes to de-risk operations before launch
- Continuous observability over tool usage and latency
Knowledge graphs & Graph RAG
Knowledge foundations that fuse structured graphs and vector recall to keep answers truthful and explainable.
- Graph ETL pipelines ingesting documents, APIs, and SQL
- Contextual retrieval combining embeddings with causal links
- Lineage tracing so stakeholders trust every response
Causal reasoning systems
Proprietary causal inference frameworks to model interventions, counterfactuals, and impact pathways.
- BLiP-inspired cause-effect detection over heterogeneous data
- Do-calculus engines to simulate decisions before rollout
- Responsible governance into every experiment
Delivery-grade product craft
Design, full-stack engineering, and MLOps working as one sprint unit to ship usable software, fast.
- Design systems with neon futurism and a11y baked in
- Composable microfrontends and API-first backend
- CI/CD pipelines with deterministic IaC provisioning
Governance & assurance
Risk frameworks and policy automation so innovation clears legal, compliance, and IT with confidence.
- Red teaming, jailbreak sweeps, and threat modeling
- Data residency, anonymization, and access controls
- Executive reporting dashboards for ongoing oversight
A four-phase relay that converts ideas into intelligent products
Every engagement runs through a calibrated relay—no silos, no hand-offs lost in translation, just a consistent rhythm from discovery to go-live.
Signal Sprint
We interrogate your data, users, and domain constraints, then blueprint an AI proposition your stakeholders will buy in on.
Architecture Lab
Design explorations, UX spikes, and model evaluations converge into a single narrative we can ship against.
Build Relay
Parallel squads assemble frontends, APIs, model services, and graph infrastructure with automated testing every step.
Lift-off & Learn
We embed telemetry, evaluate real user interactions, and tune both UX and models to compound value quickly.
Proprietary systems to jumpstart every build
Reusable assets, tooling, and playbooks that compound across clients so you inherit battle-tested foundations.
Graph RAG Fabric
Composable pipelines to hydrate dense + sparse retrieval, layer causal traces, and surface citations automatically.
Agent Ops Console
Mission control for agent orchestration with live tool usage, cost, latency, and guardrail scoring in one UI.
Model Studio
Data curation, evaluation suites, and auto-tuning flows that reduce fine-tune cycles from weeks to days.
Causal Playbooks
Scenario libraries and decision simulators to test interventions before you touch production operations.
All the right tools, orchestrated for momentum
We meet you where your architecture lives, bringing orchestration blueprints across models, infra, data, and observability.
Models & Intelligence
- OpenAI · Anthropic · Mistral
- Fine-tuning & distillation (QLoRA)
- Local inference (Ollama)
- Multi-provider orchestration with fallbacks
- Custom embeddings & retrieval
Knowledge & Retrieval
- Neo4j, Memgraph, RDF stores
- Weaviate, Pinecone, LanceDB
- Custom embedding distillation
- Causal graph compilers
- Temporal event stores
Engineering & Ops
- Next.js 16 — App Router, static & serverless
- LangGraph & Autogen
- Temporal, Airflow, Dagster
- Terraform + Pulumi IaC
- Great Expectations, Evidently
Built for scale, proven in production
From fintech to media, explore the applications TCKC has engineered — public access opens on 1 September 2026, with iArtiBot following on the 8th.
monassistant.pro
Three specialised French AI assistants: accounting, labour law, taxes — every answer cited.
A family of domain assistants that answer by voice or text and cite the official reference — chart-of-accounts number, statute article, tax-return box. Free questions at signup, then a 200-question pack with no subscription.
monassistant.pro
Three specialised French AI assistants: accounting, labour law, taxes — every answer cited.
monassistant.proFEC.NEWS
A French FEC file or tax-package PDF → full financial diagnosis, 100% deterministic numbers.
fec.newsBraquage Six
French GTA 6 news, verified in-game — news, guides and glitches with a clear status.
braquagesix.friArtiBot
You make the move, your bot does the rest — the tradesperson's assistant, paid per task.
iartibot.frdjian-gutenberg.com
Redesign for a reference astrologer: bilingual showcase, consultations, school, works.
djian-gutenberg.comHeliosForge
Stellar engineering argued through reproducible simulations — Dyson swarm, star lifting.
github.com/zepef/HeliosForgeParles au PCG
Voice-enabled French accounting assistant
parles-au-pcg.comParles aux Impôts
Voice tax assistant — 2026 French return
parles-aux-impots.comProof, not promises
We prefer verifiable guarantees to testimonials: here is what our products promise — and hold — by construction.
“No figure ever comes out of a language model: a deterministic rules engine computes, the AI explains. Same files in, same results out, to the cent.”
Design principle — FEC.NEWS“Every answer cites its official reference: chart-of-accounts number, statute article, tax-return box. And every accounting entry is verified before display.”
Product guarantees — monassistant.proWe’re lining up our next wave of case stories. Ready to become the next headline?
Book the discovery sprintLet’s architect your causal AI advantage
Drop us the problem space, timeline, and data constraints. We’ll respond within one business day with a recommended sprint plan.
