Senior Software Engineer building across data platforms, fintech, and climate tech at scale.
I build AI automation and workflow systems. Agentic pipelines, retrieval-augmented assistants, and integrations that compound over time.
STACK ▸
TypeScriptNestJSPythonFastAPIDjango
ReactNext.jsPostgreSQLKafkaRabbitMQ
GCPAzureDatabricksKubernetesTerraform
Delta LakeAirflowWeb3.js
TypeScriptNestJSPythonFastAPIDjango
ReactNext.jsPostgreSQLKafkaRabbitMQ
GCPAzureDatabricksKubernetesTerraform
Hackathons & build3s
Watch the buildsProjects
All · Filter
Full Stack
DEXPAY · DeFi P2P Platform
Decentralised peer-to-peer platform for buying and selling crypto via non-custodial wallets (Metamask / Trust Wallet) using smart contracts. Built with Node.js, NestJS microservices and Next.js. Converts crypto to fiat through a clean P2P flow.
Next.jsTypeScriptNestJSMicroservicesMongoDBDockerReduxTailwind
Full Stack
Team Coach
IT Services and Consulting platform integrating collaboration, operations, community and portfolio management in one place. Built with Node.js and React to help teams improve operational productivity and accelerate startup growth.
ReactNode.jsTypeScriptPostgreSQLDockerReduxMaterialUI
Data project · MapReduce
Data Science
Product Rating Insights with MapReduce
Analyses customer review data using Cloudera and Hive. MapReduce aggregates 5-star and 1-star ratings per product; visualisations surface top and bottom performers so businesses can make data-driven product, marketing and CX decisions.
PythonMapReduceHDFSHiveSQL
Credit risk · Ensemble ML
Data Science
CreditCompass
Predicts loan approval status from customer loan application data using Decision Trees and SVM, ensembled to improve accuracy. Identifies key drivers of loan eligibility and quantifies risk per application, designed to help lenders make better decisions while mitigating risk.
PythonScikit-learnEnsembleSeabornPandasNumPy
Machine Learning
Human Activity Recognition
Trains a model on the WISDM dataset to classify physical movements like walking, jogging, sitting, and standing from smartphone and smartwatch sensors. End-to-end pipeline: preprocessing, feature engineering, model training and evaluation.
PythonScikit-learnSeabornPandasNumPy
NLP · Deep Learning
Machine Learning
Science-Focused Chatbot
Conversational agent that answers science questions. Built in Python with NLTK and TensorFlow for an engaging, informative conversational experience tuned to a science domain.
PythonTensorFlowNLTKNumPyDeep Learning
Sentiment · NLP
Data Science
Review2Revenue
Sentiment-driven analysis of product reviews using NLTK, VADER and deep learning. Surfaces emerging trends and investment opportunities per product category, translated into clear visualisations for non-technical stakeholders.
PythonNLTKScikit-learnVADERSeabornPandasNumPy
Credit risk · Model comparison
Machine Learning
Credit Card Default Prediction
Comparative study of ML algorithms for credit card default prediction. Benchmarks accuracy, precision and recall across models to identify the strongest performer for risk forecasting, informing credit approval, interest-rate, and credit-limit decisions.
PythonScikit-learnSeabornPandasNumPy
Data Science
Human Activity Recognition · Visualisation
Unveils patterns in motion through visualisation: boxplots, scatter plots and line charts transforming raw smartphone/smartwatch sensor data into actionable insights for researchers, clinicians and fitness practitioners.
PythonMatplotlibScikit-learnSeabornPandasNumPy
Machine Learning · AI Agents
RAG-Assisted Data Analyst
A data-analyst agent powered by Retrieval-Augmented Generation that combines a pre-trained language model with real-time retrieval so responses stay accurate, contextual and up to date. Uses Cohere Command R+ for generation and Langchain for tool orchestration.
PythonRAGLangChainCohereTavily
Work experience
2020 · PresentCarbon Instinct ↗
Senior Software Engineer, Data
Owning the ETL backbone, Node-Edge graph calculation engine, and multi-level emissions analytics for an
enterprise climate platform on GCP + Kubernetes.
Jan 2025 · Present
London, UK
London, UK
Dexpay Inc. ↗
Data & Software Engineer
Designed the Azure data architecture and event-driven payment pipelines powering a decentralised finance
product, with observability, security, and consistency baked in.
Nov 2023 · Dec 2024
TeamCoach ↗
Software Engineer
Led the freemium tier launch, feature-flag platform, and a full state-management migration from
React-Redux to TanStack Query.
May 2022 · Oct 2023
Interswitch ↗
Software Engineer
Rebuilt the Extraswitch reporting engine, shipped a real-time multi-region payments frontend, and
designed PCI-DSS-compliant payment integrations.
Jan 2020 · Apr 2022
A few bits about me
AboutI enjoy weekends of Formula One racing, I take part in marathons, and I love hiking and traveling. I also get lost in documentaries and reading about random topics that have nothing to do with my day to day.
I'm someone who looks for ways to improve things. When a process is clunky, I'm usually thinking about the simplest way to make it work better.
Let's build
something that matters.
something that matters.