Skip to content
Based in London · Open to AI, data, analytics and research opportunities

Applied AI that makes complex decisions easier to understand.

I'm Akmam Hasan, an Applied AI and data professional with an MSc in Applied Artificial Intelligence. I build explainable machine-learning, NLP and data solutions across risk analytics, scientific forecasting and operational decision support.

My work combines careful data preparation, model comparison, explainability and clear communication—so that technical results can support practical human decisions.

Explore my work Download CV
GitHub LinkedIn Email
Python · SQL · Machine Learning · NLP · Explainable AI · Data Visualisation
Akmam Hasan portrait
Akmam Hasan
London, UK
Applied AI / Data / Research
Evidence at a glance

Verified figures from the research and project work.

542
Companies in multimodal ESG research
≈0.71
Weighted F1 in a multimodal classification experiment
98%
Anomaly-detection accuracy in the project setting
1
Peer-reviewed Springer publication

Metrics reflect the datasets, experimental settings and evaluation methods of the respective academic projects.

Selected work

Projects built around real decisions, not just algorithms.

These case studies show how I move from problem definition and data preparation to modelling, validation, interpretation and practical communication.

Multimodal AI · NLP · Explainability

LenseESG — Explainable Multimodal ESG Risk Intelligence

An end-to-end research pipeline combining structured ESG indicators, company disclosures and news-derived signals to create more interpretable company-risk assessments.

542 companies
Structured, text-only and multimodal experiments
Approximately 0.71 weighted F1
SHAP and transformer-attention analysis
PythonPandasscikit-learnLightGBMBERTESG-BERT
View case study
Scientific ML · Time Series · Published Research

Deep Learning for Cyclone Track Forecasting

A Springer-published comparative study of LSTM, GRU and ConvLSTM models for forecasting cyclone trajectories from historical Bay of Bengal data.

Peer-reviewed Springer chapter
Sequential trajectory modelling
Multiple error measures
Practical disaster-risk relevance
PythonLSTMGRUConvLSTMTime-series evaluation
View case study View Springer publication
Time Series · Monitoring · Edge Analytics

Streaming Anomaly Detection for Cyber-Physical Systems

A complete monitoring workflow that connected simulated sensor data, anomaly-detection models, time-series storage and live dashboard reporting.

98% project accuracy
Four-model comparison
Streaming-data workflow
Operational Grafana dashboard
MATLABSimulinkNode-REDInfluxDBGrafanaRandom Forest
View case study
SQL · PostgreSQL · Retool · Data Operations

Data Reconciliation and Operations Monitoring Dashboard

A sanitised operational analytics case study that reconciled records across two source systems, identified missing encounters and turned exceptions into an actionable dashboard.

25,956 source rows
6,612 valid encounters
6,330 matched
282 missing
PostgreSQLSQLRetoolData cleaningReconciliationKPI reporting
View case study
View all projects
Working approach

From an unclear question to an evidence-based solution.

A consistent five-stage workflow keeps modelling honest, communicates limits clearly, and turns proofs of concept into useful outputs.

  1. 01
    Frame

    Define the real problem, affected users and what a useful outcome would look like.

  2. 02
    Prepare

    Clean and validate data while checking missingness, inconsistent formats, class imbalance and leakage risks.

  3. 03
    Compare

    Establish baselines, test appropriate methods and choose metrics that reflect the real use case.

  4. 04
    Explain

    Interpret outputs, limitations and uncertainty so that results remain understandable to technical and non-technical audiences.

  5. 05
    Deliver

    Translate the proof of concept into dashboards, workflows, monitoring needs and practical next steps.

Capabilities

Technical depth with business context.

Skills grouped by how I actually use them in projects — modelling, language, explainability, analytics and delivery.

Machine learning

  • Regression and classification
  • Feature engineering
  • Model benchmarking
  • Hyperparameter tuning
  • Validation and error analysis
  • Class-imbalance handling
  • Leakage detection

NLP and deep learning

  • BERT and ESG-BERT
  • Tokenisation
  • CLS embeddings
  • Text classification
  • Early multimodal fusion
  • LSTM
  • GRU
  • ConvLSTM
  • CNN workflows

Explainability and responsible AI

  • SHAP
  • Feature attribution
  • Transformer attention visualisation
  • Interpretable model reporting
  • Human oversight
  • Limitation and validity analysis

Data and analytics

  • Python
  • SQL and PostgreSQL
  • Pandas
  • NumPy
  • scikit-learn
  • LightGBM
  • Excel
  • Tableau
  • Grafana
  • Retool

Delivery and systems

  • Git and GitHub
  • Jupyter and Google Colab
  • Ubuntu
  • OpenStack
  • Node-RED
  • InfluxDB
  • MATLAB and Simulink
  • APIs
  • Requirements analysis
  • SRS documentation
  • Stakeholder communication
Published research

Deep learning for cyclone trajectory forecasting.

The study compared recurrent deep-learning architectures for predicting cyclone trajectories using historical Bay of Bengal data. Performance was evaluated using both statistical metrics and geographical distance error so that model behaviour could be interpreted in a practical forecasting context.

Citation

Majid, R., Hasan, A., Mostarin, S., Alam, K.R., & Rahman, R.M. (2024). "A Deep Learning-Based Study for Cyclone Track Forecasting: Comparative Analysis Using Historical Data from the Bay of Bengal." In Networking and Parallel/Distributed Computing Systems, Studies in Computational Intelligence, Vol. 1125. Springer, Cham.

DOI: 10.1007/978-3-031-53274-0_1
Read publication Explore research
Forecast vs observation

Predicted vs observed cyclone latitude (°N)

Deep learning trajectory model tracks observed Bay of Bengal cyclone path over a 72-hour horizon.

Experience

Experience across technology, analysis and frontline operations.

  1. April 2024 – Present

    Customer Advisor — B&Q

    London, United Kingdom

    Work directly with customers and store teams to solve product, order, collection, return and service issues in a fast-paced retail environment. The role has strengthened practical problem-solving, service recovery, operational awareness and communication with non-technical audiences.

  2. July 2023 – October 2023

    Software Engineer Intern — Echologyx Ltd

    Supported website optimisation and A/B-testing activities by implementing frontend variants with JavaScript, HTML and CSS, checking quality before release and collaborating with technical and product stakeholders.

  3. April 2021 – May 2022

    Requirement Analyst — Generic Solutions

    Bangladesh

    Gathered stakeholder requirements, mapped workflows, prepared software-requirement documentation and translated business needs into structured technical tasks.

  4. June 2022 – February 2023

    Project Management Intern — Youth Opportunities

    Coordinated webinars, outreach, recruitment support, communication and follow-up activities across remote stakeholders.

Read my full story
Education

Foundations across applied AI and computer science.

2023–2025

MSc Applied Artificial Intelligence

London South Bank University · London, United Kingdom
Merit
Machine LearningDeep LearningPython Programming for AI and VisualisationResearch MethodsIndustrial Cyber-Physical SystemsFuture Internet Technologies

Multimodal ESG risk prediction using structured ESG indicators, corporate and news text, transformer representations, model comparison and explainability.

Graduated 2023

BSc Computer Science and Engineering

North South University · Bangladesh
CGPA 3.61/4.00 · Distinction
Specialisation: Artificial Intelligence

AI-focused thesis research led to a Springer-published chapter on deep-learning-based cyclone track forecasting.

Get in touch

Looking for someone who can connect AI, data and practical decision-making?

I am interested in opportunities involving applied AI, data analytics, machine learning, NLP, explainability, data operations and research.

Email Akmam View LinkedInExplore projects