01
Computer Vision
Production-grade vision for medical, industrial and earth-observation. Detection, segmentation, super-resolution and real-time inference on the edge.
Learn more
ARCHEON SOLUTIONS · SF / ZÜRICH · EST. 2026
N 47°22′ · W 122°25′SYSTEMS ONLINE
Research-grade AI,
shipped into production.
We are an AI research and engineering studio for tier-1 banks, insurers and industrial leaders. Built by scientists with world-record inference, published medical research, and operator experience at NVIDIA, Siemens Healthineers and JP Morgan — not prompt engineers.
COMPUTER VISIONLLM OPTIMIZATIONHEALTHCAREFINANCER&D
Benchmark
World record
DeepSeek V3 inference throughput — open-source, reproducible.
Now accepting Q3 engagements
Scroll
Positioning
A studio, not an agency.
Scientists, not slideware.
Most “AI consultancies” sell prompt engineering and dashboards. We build the systems banks, hospitals and industrial operators run in production — measured in latency, cost and accuracy, not in slide decks.
Archeon is a small, senior team of French and American AI scientists based in San Francisco, Zürich, Paris, and Dubai. We embed with your engineers, ship working systems, transfer knowledge — and leave behind infrastructure your team owns.
01 / Proof
World record
DeepSeek V3 inference throughput — published, reproducible.
02 / Proof
Peer-reviewed
Medical imaging research with Siemens Healthineers and UCL.
03 / Proof
Operators
Built systems at NVIDIA, JP Morgan, Siemens Healthineers, Avian.
What we build.
Five engineering practices, one team. We sequence them per engagement — vision in production, language at scale, ML infra that survives audit.
02
LLM Engineering
On-prem and air-gapped LLMs. Custom kernels, quantization, speculative decoding — the work that took us to a world record on DeepSeek V3.
Learn more
03
Healthcare AI
Clinical-grade models with regulatory awareness. Surgical assistance, imaging and diagnostics that meet hospital reliability bars.
Learn more
04
Finance AI
Risk, alpha research and document intelligence for banks and asset managers. Models that survive compliance, audit and stress.
Learn more
05
Research & Development
Frontier research turned into deployable systems. We co-author with Berkeley, Harvard and UCL — and ship what works to your stack.
Learn more
Verticals we know cold.
We work where mistakes are expensive — regulated industries with measurable AI ROI. Pick a vertical to see how we engage.
Vertical brief
Banking & Finance
We build risk engines, on-prem LLM stacks and document-intelligence systems for tier-1 banks and asset managers. Auditable from training data to inference, deployed inside your perimeter — no third-party model APIs.
60%
Manual review reduction
< 50ms
P99 inference latency
0
Data leaving perimeter
Selected partners
JP MorganAvian
Selected work.
Engagements where the difference between research and production was measurable — in latency, in dollars, in patient outcomes.
Trained inside the systems we now replace.
Our team built and ran AI inside the institutions our clients trust. We come back with the playbook.
01 / Operator
NVIDIA
AI & Accelerated Computing
Inference optimization, custom CUDA kernels and accelerated training pipelines.
- LLM inference optimization
- CUDA + TensorRT pipelines
- Distributed training infra
02 / Operator
Siemens Healthineers
Medical AI
Clinical imaging, diagnostic models and registration systems shipped into hospitals.
- Diffusion-based CT super-resolution
- Real-time surgical vision
- Clinical integration & validation
03 / Operator
JP Morgan
Financial AI
On-prem LLM tooling and document intelligence inside a tier-1 banking perimeter.
- On-prem inference stacks
- Risk & compliance modeling
- Audit-ready ML systems
A studio of scientists.
Archeon is a small, senior team of AI researchers and engineers based in San Francisco, Zürich, Paris, and Dubai. We embed with your operators, ship working systems, and leave behind infrastructure and people your team owns.
0
World record · DeepSeek V3
0
PhDs & research alumni
0
Tier-1 enterprise engagements
0
Offices · SF · Zürich · Paris · Dubai
The people on your engagement
Leadership
Senior practitioners, no junior consultants. The names on the proposal are the people in your meetings.
PLD
Pierre-Louis Delcroix
LLM Engineering
Production ML across healthcare and fintech. World-record LLM inference at 303 tokens/sec on DeepSeek R1. ML research on diffusion models for CT imaging at Siemens Healthineers.
TB
Théo Bonzi
Computer Vision Engineering
Ex-UC Berkeley research, published in astronomical ML. Edge OCR architect — 10× faster inference on CPU-only hardware. Real-time medical ultrasound AI and generative vision from R&D to production.
AW
August Weinbren
Medical ML Engineering
Doctoral researcher at UCL. ML engineer on radiotherapy optimization and medical image registration at Siemens Healthineers. LLM inference specialist with CUDA, TensorRT, and speculative decoding.
JJ
Jessup Jong
Geospatial AI & Remote Sensing
Deep learning engineer leading applied research and performance benchmarking. Computer vision on satellite imagery for geospatial asset tracking. Former ML engineer at Upstage building LLM systems.
Research network
Active collaborations with research groups at Berkeley, Harvard and UCL.
What partners say01 / 03
“Archeon's hyperspectral analysis pipeline transformed our environmental monitoring program. Their team delivered accuracy levels we didn't think were achievable with satellite data alone.”
Start with a working session.
Bring a real problem. We bring a senior engineer and a scientist. 30 minutes — no slides, no qualifying call.
Direct line
contact@archeon.solutions
San Francisco
Engineering · Research
Zürich
Client & Operations
Paris
Europe · Research
Dubai
Middle East · Operations
Available globally — engagements delivered on-prem, hybrid or air-gapped as your compliance requires.