Skip to content

Open to full-time roles · India & international

Mohammed Adan · AI Engineer · Computer Vision · Deep LearningI build deep-learning models and the applications that put them to work.

AI engineer specialising in computer vision. I take models from raw image data to explainable, production-oriented applications, currently as a Junior Computer Vision Engineer at Unicorn Medics.

Explainability explorer

Chest X-ray, scan 1, as uploaded to LungVision.Scan 1 with the Grad-CAM heatmap from model C; the highest attention lies over both lung fields. Colours corrected.
Scan
Model
lowerrelative attentionhigher

Real LungVision output. Drag to compare the scan with the Grad-CAM heatmap from one of its four ensemble models (colours corrected for a channel-order bug in the app). A heatmap shows where a model looked, not whether it was right.

How to read this

Grad-CAM colours where a network’s last convolutional features pushed its prediction: red is higher relative attention, blue lower. Each heatmap is scaled per image, so colours compare regions within one image, not across images.

Models A–D are the four ensemble networks’ outputs in the order the app saved them. The app does not record which architecture made which file, so the same letter on two scans may be different networks.

LIME marks superpixels that support (green) or oppose (red) the prediction. Its four runs per scan rarely overlap, so treat any single run with caution.

Blend fades the app’s saved overlay over the original; nothing is recomputed. These are research outputs, not clinical findings. Full analysis.

  1. 01Model

    Prepare image data, train and compare models, evaluate honestly, explain predictions.

  2. 02Application

    REST APIs, databases, authentication and reporting around the model.

  3. 03Deployment

    Containerised services with health checks, tests and repeatable configuration.

Mysore, India · open to relocation · B.E. CSE (AI), VTU 2026

Selected work

Projects that go from data to a working product.

Labelled for what each project is (professional, academic or independent). Every result names its source.

WEB APPLICATION · CONTAINERISEDRadiographClassifierGrad-CAMReportCase history

Professional· Unicorn MedicsJun 2026 – present

Clinical X-ray AI · Unicorn Medics

My professional work as a Junior Computer Vision Engineer at Unicorn Medics: owning the path from X-ray data and model development to an explainable, containerised application used for clinical review.

  • Python
  • TensorFlow / Keras
  • OpenCV
  • Grad-CAM

Schematic diagram of the system.

A chest X-ray uploaded to the LungVision application.The same chest X-ray with the Grad-CAM heatmap from model C of the LungVision ensemble, concentrated over both lung fields (colours corrected).

Academic· Team of 4Mar 2025 – Jan 2026

LungVision

A final-year team project that combines chest X-rays, CT scans and clinical metadata for early lung cancer detection, with four-CNN ensembles per imaging modality and Grad-CAM and LIME explanations for every prediction.

1st place, final-year project evaluation

  • PyTorch
  • Flask
  • Grad-CAM
  • LIME

Real LungVision output: the same scan and model, colours corrected.

VALIDATION ACCURACY · 20 EPOCHS93%95%97%besttestedTEST · n = 231103101117DRNO DR95.2% ACCURACY
From the notebook’s saved outputs

Independent project2025

Diabetic retinopathy screening

A ResNet-18 transfer-learning model that screens retinal fundus images for diabetic retinopathy, served through a Flask app with a confidence score and a Grad-CAM heatmap, and audited honestly afterwards.

95.2%test accuracy on 231 held-out images (final-epoch model)Source: Repository notebook

  • PyTorch
  • ResNet-18
  • Grad-CAM
  • OpenCV

Plotted from the notebook’s saved outputs: validation accuracy per epoch and the test-set confusion matrix.

ARIMA · PROPHET · LSTM · TRANSFORMERobservedforecastthreshold

Independent prototype2026

Network forecasting

An independent prototype that forecasts network throughput, latency and packet loss with statistical and deep-learning models, raises early warnings and drafts incident reports. Built as a tested Python package and developed on synthetic perfSONAR-style data.

  • Python
  • PyTorch
  • statsmodels
  • Prophet

Schematic diagram of the forecasting and alerting flow.

All work, including 2 more projects

Experience

Where I’ve built and shipped.

  1. – Present

    Professional

    Remote

    Junior Computer Vision Engineer

    Unicorn Medics

    Computer vision engineering on a medical-imaging product, covering the path from radiograph data to a deployed, explainable application.

    • Prepare and analyse X-ray image datasets: cleaning, preprocessing, exploratory data analysis and annotation workflows for detection tasks.
    • Train and compare deep-learning models with transfer learning, and review model outputs with clinicians.
    • Integrate Grad-CAM explanations so every prediction can be inspected visually.
    • Build the web application around the model: REST APIs, database-backed case history, role-based access and report generation.
    • Containerise the application with health checks, automated tests and backup and restore procedures.
  2. –

    Fellowship / internship

    Remote

    AI Intern

    LearnersByte · Bharat Unnati AI Fellowship (AICTE-approved)

    A 17-week applied AI fellowship programme.

    • Built 10+ projects with Python, Flask, REST APIs, TensorFlow, PyTorch, scikit-learn and OpenCV.
  3. –

    Contract

    Remote · USA-based company

    React Native Developer

    GoMind AI LLC

    Contract work on a production React Native app, across the interface and its backend data.

    • Resolved 10+ UI defects across splash screen, button and dropdown components, improving app stability.
    • Automated cleanup of inactive user data (30+ days), reducing database load by 20%.
    • Optimised the logging pipeline, cutting unnecessary logs by 30%.

Achievements

Recognition, teaching and communication.

  • 1stplace

    First Place — Final-Year Project Evaluation

    LungVision, our multimodal lung cancer detection system, won first place in the department's final-year project evaluation.

    Recognised for technical excellence, innovation and real-world impact. Also presented at Techtatwa 2K25.

    Read the LungVision case study
  • 20+projects assessed

    Date: Apr 2026

    Hackathon Evaluator

    Served as an ML domain evaluator for the Xypher Hackathon.

    Assessed 20+ projects for technical innovation and real-world applicability.

  • 100+peers, 2 workshops

    Date: Jul 2026

    Workshop Facilitator

    Conducted 2 hands-on ML/AI workshops for 100+ peers.

    Covered model building, deployment and career guidance in AI.

  • 4languages

    Multilingual Communicator

    Fluent in English, Kannada, Urdu and Hindi.

    Comfortable collaborating across cultures and keeping team communication inclusive.

About

Trusted, not only accurate.

I build computer vision systems end to end, and I care whether they can be trusted, not only whether they score well.

My route into AI engineering ran through medical imaging: retinal fundus photographs, brain MRI, chest X-rays and CT scans. Those projects taught me that a good validation score is only the start. The data has to be clean, the evaluation has to be honest, and the model has to explain itself to the people who rely on it.

Today I am a Junior Computer Vision Engineer at Unicorn Medics, and I am looking for full-time roles where I can own a model together with the software that puts it to work.

Full story, skills and education

Skills at a glance

Deep learning
PyTorch · TensorFlow / Keras · scikit-learn · CNNs · Transfer learning
Computer vision
OpenCV · EfficientNet · ResNet · DenseNet · MobileNet · VGG · YOLO annotation workflows · X-ray, CT, MRI & fundus preprocessing · DICOM
Evaluation & explainability
Grad-CAM · LIME · ROC-AUC · F1-score · Cohen’s kappa
Backend & data
Flask · FastAPI · REST APIs · SQLAlchemy · Authentication & role-based access
Deployment & tooling
Docker · Docker Compose · Gunicorn · Health checks · Automated testing (pytest)

Education

B.E. Computer Science & Engineering (Artificial Intelligence)

Maharaja Institute of Technology Mysore · Visvesvaraya Technological University (VTU) · Dec 2022 – Jun 2026 · First Class with Distinction

Google Data Analytics Professional Certificate · IBM Data Science Professional Certificate · IELTS · Overall band 7.5

Contact

Hiring for AI or computer vision engineering? Let’s talk.

Open to full-time engineering roles across India and internationally. Available to relocate immediately; visa sponsorship required where applicable.