Skip to content
Open to AI/ML roles & research collaborations
Based in Linz, Austria

Ahmed
Mohammed

Computer Vision · Diffusion Models · OOD Detection

AI/ML engineer. I build ML systems I can stand behind — verified error bars in research, eval harnesses in production.

View work →Get in touchSchedule meetingDownload CV
4+
Years in AI/ML
99.03%
OOD Detection AUROC
+6.5pp
Gain over baseline (stable)

99.03% AUROC = CIFAR-10 thesis (binary, single-class airplane-vs-rest, 3-seed mean) · 0.8673 AUROC = industrial FTI_Zer0P 5-fold baseline · +6.5pp = gain from separation loss

JKU Linz, Austria
Supervised by Prof. Sepp Hochreiter
PROFACTOR / FTI_Zer0P
99.03% AUROC thesis result (binary, single-class)
AI QC platform shipped to production
// 01 - About
Ahmed Mohammed

AI/ML Engineer - research that ships.

Transitioned from mechatronics engineering to AI/ML - bringing a hardware and systems perspective to computer vision and production ML.

I am an AI/ML engineer who completed an M.Sc. in Artificial Intelligence at JKU Linz (graduated Sep 2026), working under Prof. Sepp Hochreiter.

My work spans the full stack: from class-conditional separation loss for diffusion-based OOD detection to industrial computer vision pipelines evaluated under rigorous cross-validation. 99.03% +/- 0.07% AUROC (binary airplane-vs-rest, 3-seed mean) · 0.8673 AUROC (FTI_Zer0P 5-fold).

I also build and ship — most recently Sihem, an LLM personal-mentor assistant, and before that Faultrix, an AI quality-control platform I took from zero to production on my own. Both taught me that production reliability is its own kind of rigor.

Location
Linz, Austria
Supervisor
Prof. Sepp Hochreiter
Current work
Sihem — LLM assistant (live beta)
Open to
AI/ML Roles - Research Collaborations
Education
M.Sc. in Artificial Intelligence
Oct 2024 - Sep 2026
Johannes Kepler University Linz
Thesis: conditional diffusion models for OOD detection - 99.03% +/- 0.07% average AUROC (binary airplane-vs-rest, 3-seed mean).
B.Sc. in Mechatronics Engineering
Feb 2015 - Jan 2018
Eastern Mediterranean University, Cyprus
Thesis: SCARA robotic system for dynamic object tracking.
Skills & Stack
Deep Learning
PyTorchPyTorch LightningDiffusion ModelsCNNsTransformersRNNsLLMs
Computer Vision
OOD DetectionObject DetectionDefect DetectionImage Classification
Research & Experimentation
HydraExperiment DesignAblation StudiesAUROC/FPR95Monte CarloCross-Validation
ML Engineering
End-to-End PipelinesLLM InferencePrompt EngineeringOpenAI API
Infrastructure
DockerGitLinuxREST APIsTypeScriptNext.jsCUDAPythonOpenCV
AI Tools
GitHub CopilotChatGPTWeights & Biases
Languages
Arabic
Native
English
B2 - Professional
German
B1 CEFR - improving
// 02 - Projects

Research that ships.

Current project · Live beta
Sihem — AI Mentor Assistant
A proactive personal-mentor assistant on Telegram that plans your week, sends morning briefings and evening reviews, and fires smart check-in triggers. Built on Supabase edge functions (Deno/TS) with a generic activity model (workouts, medication, study sessions, custom habits — no schema change per type), a deterministic generators→arbiter→composer decision engine for adaptive nudging, pgvector long-term memory with RAG, multi-provider LLM routing (GLM/Gemini/Groq) with provider failover, multimodal meal-photo vision, TTS voice reminders (Cloudflare MeloTTS), Google Calendar sync, optional Alexa Echo integration, invite-code gated beta access, and an ops dashboard with cost rollups. Now in beta — try it on Telegram at @sihem_ai_bot.
Deno / TSSupabase Edge Functionspgvector RAGLLM Routingpg_cronTelegram
Telegram bot · PWA · pgvector memory
Master's Thesis - JKU Linz
OOD Detection Framework
Conditional diffusion models as generative classifiers achieving 99.03% +/- 0.07% average AUROC on a within-CIFAR airplane-vs-rest binary split (single ID class; not comparable to multi-class CIFAR-10 OOD benchmarks). Introduced class-conditional separation loss improving average performance by 6.5 percentage points with high seed stability.
99.03% +/- 0.07% Avg AUROC
Master's Thesis Track 2
InkjetOOD Defect Detection
YOLO+CDM quality classification pipeline evaluated on the public FTI_Zer0P benchmark under strict 5-fold cross-validation.
0.8673 AUROC (5-Fold CV)
Previous project · 2025-2026
Faultrix
AI-powered construction quality control SaaS. Analyzed building photos and generated ONORM-compliant reports in under 1 minute - SHA-256 evidence chain, DSGVO compliant, AES-256 encryption. Built solo from zero to production.
<1 min per report
Next.jsConvexOpenAIClerkCloudflare R2Stripe
Karunya University · BCI Research
EEG Signal Classification
LSTM, Bi-LSTM, and GRU architectures for motor imagery classification in brain-computer interface applications.
// 03 - Proof

Selected evidence.

Public artifacts, reports, and repositories behind the portfolio claims.

Proof
Master thesis PDF
99.03% +/- 0.07% AUROC
Open →
Proof
Industrial OOD baseline (thesis)
0.8673 +/- 0.0230 AUROC · FTI_Zer0P 5-fold
Open →
Case study
Early technical report (2024)
Negative result — motivated the thesis approach
Open →
Proof
DiffusionOOD repository
CIFAR-10 OOD implementation
Open →
Proof
InkjetOOD repository
Industrial transfer codebase
Open →
Proof
Hugging Face model weights
Public checkpoints and artifacts
Open →
Proof
Research poster
Visual summary of the thesis
Open →
Case study
DiffusionOOD case study
Problem, method, results, lessons
Open →
Case study
InkjetOOD case study
Industrial transfer, results, and lessons
Open →
Proof
Sihem — live beta
Running system — Telegram bot and PWA, open to try
Open →
Case study
Faultrix case study
Research-to-product build story
Open →
// 04 - Experience

Where I've worked.

Jul 2026 Present
AI/ML Engineer
Sihem · Linz, Austria
Building an LLM personal-mentor assistant on Supabase edge functions (Deno/TS) with pgvector RAG memory, multi-provider LLM routing with failover, and a deterministic decision engine for adaptive nudging.
Live beta - @sihem_ai_bot
Jul 2025 Jul 2026
AI/ML & Full-Stack Engineer
Faultrix · Linz, Austria
Built an AI construction analysis platform solo from zero to production with Python, Next.js, Convex, OpenAI API, Docker, Clerk, Cloudflare R2, and Stripe.
Shipped - faultrix.com
Dec 2024 Jul 2026
ML Researcher
JKU Machine Learning Institute · Linz
Master thesis on conditional diffusion models as generative classifiers for out-of-distribution detection under Prof. Sepp Hochreiter.
Apr 2024 Nov 2024
Machine Vision Researcher
PROFACTOR GmbH · Steyr, Austria
YOLO + diffusion pipeline for industrial defect detection in zero-defect inkjet printing on building components.
98.4% (threshold-dep., production)
Aug 2023 Oct 2023
AI Research Intern
Karunya University · India (Remote)
RNN/CNN architectures for EEG motor imagery classification with hyperparameter optimization.
Jan 2021 Present
AI & Programming Tutor
Freelance
Tailored training in Python, ML, and deep learning, plus mentoring on portfolio projects and GitHub practice.
// 05 - Research

Reports & Thesis

Problem
How can a model recognize when an image does not belong, instead of acting overconfident?
Contribution
A separation loss that forces the model s competing class explanations to move apart.
Result
99.03% +/- 0.07% AUROC on CIFAR-10 airplane-vs-rest (binary, single ID class, 3-seed mean), with a +6.5pp gain over the non-separated baseline.
Why it matters
The score became not only higher, but dramatically more stable across seeds and easier to trust.
Master's Thesis (Track 1) - JKU Linz - 2026
Conditional Diffusion Models as Generative Classifiers for Out-of-Distribution Detection in Inkjet Print Quality Control
A diffusion model compares how well different class explanations reconstruct an image. The new separation loss pushes those explanations apart, making the anomaly score clearer and more stable. 99.03% +/- 0.07% AUROC, +6.5pp gain, baseline 92.52% +/- 11.07%, external OOD 90.50%-96.97%.
Supervisor: Prof. Sepp Hochreiter - Assistant: Claus Hofmann, MSc - JKU Linz - 2026
99.03%
AUROC - CIFAR-10 Average
+/-0.07% variance (3 seeds)
Separation Loss Weight Sweep
Separation weight sweep: best stability at lambda=0.02
Operating Threshold Calibration
Score distributions at calibrated operating threshold
Per-Timestep Reconstruction Error
Reconstruction error gap peaks at intermediate noise
Novel contribution - separation loss
Plain idea: train the model to make its two explanations visibly different. The gap becomes the anomaly signal.
1Learn to denoiseReconstruct images under each class condition.
2Push explanations apartIncrease the distance between conditional predictions.
3Score the gapUse the reconstruction-error difference as the OOD signal.
Total loss=denoising loss+lambda * separation loss
Latest thesis result: lambda = 0.02 reached 99.03% +/- 0.07% AUROC, a +6.5pp gain over no separation loss.
View Full Thesis (PDF) ->
Master's Thesis (Track 2) - JKU Linz - 2026
Rigorous Cross-Domain Transfer & FTI_Zer0P Benchmark
The public YOLO+CDM print-quality pipeline was evaluated on FTI_Zer0P with strict 5-fold cross-validation, showing where separation loss transfers and where small manufacturing textures behave differently. 0.8673 +/- 0.0230 AUROC.
Public Benchmark - InkjetOOD Codebase - FTI_Zer0P Public Dataset
0.8673
AUROC - 5-Fold CV Baseline
Rigorous Public Evaluation
Cross-Domain Impact of Separation Loss
Cross-domain impact: CIFAR-10 gain vs. Inkjet QC variance
Per-Feature AUROC Sweep
Per-feature AUROC across separation weights
View Early Technical Report (2024) ->Code: InkjetOOD ->HF: Model Weights ->R&D Notebooks ->
Motor Imagery Classification Based Brain-Computer Interface for Rehabilitation
Karunya University - 2023
->
Reading Thoughts Using GANs
Project Report - JKU Linz - 2023
->
// 06 - Writing

Thinking out loud.

Deep-dives on research decisions, production lessons, and the gap between papers and products.

Diffusion ModelsOOD Detection
How I Reached 99.03% AUROC on OOD Detection with Conditional Diffusion Models
My master's thesis at JKU Linz introduced class-conditional separation loss into conditional diffusion models used as generative classifiers...
14 min readRead →
LLMSaaS
Shipping Faultrix: What I Learned Building an AI SaaS From Zero
What changed when I moved from research into product building: Faultrix, an AI SaaS for construction quality control that generated ONORM-al...
9 min readRead →
Diffusion ModelsAnomaly Detection
Diffusion Models for Industrial Defect Detection at PROFACTOR GmbH
How I evaluated a public YOLO + conditional diffusion pipeline on the FTI_Zer0P benchmark under strict 5-fold cross-validation, reaching a 0...
11 min readRead →
View all posts →
// 07 - News

Tracking the field.

Curated AI news with a short take on each — with a bias toward agentic AI, where I spend most of my attention.

Agentic AIOpenAI
Introducing the Agents API
OpenAI turning the Codex harness itself into a managed API — sessions, context compaction, tool search, subagents, and a choice of hosted or...
2026-09-10Read take →
Agentic AICursor
Introducing Projects
Cursor moving the product up from chats and sessions to months-long bodies of work — a coordinator delegating to thousands of subagents and ...
2026-09-10Read take →
Agentic AICognition
Introducing SWE-2: Pushing the Pareto Frontier
SWE-2 is the strongest data point yet that open weights are becoming raw material: Cognition took Moonshot's 2.8T-parameter Kimi K3, ran its...
2026-09-10Read take →
View all news →
// 08 - Contact

Let's build something extraordinary

Open to AI/ML roles, research collaborations, and product partnerships.

GitHub →LinkedIn →Hugging Face →W&B →Schedule meeting ->Download CV ->
ahmed.mo.0595@gmail.com