Teddy Ambona

Teddy AMBONA Thailand - (Whatsapp) - (Thai mobile) -Linkedin profile github.com/teddy-ambona Versatile Python Engineer with 6Y+ of experience in SaaS, cloud/web development, automation, designing battle-tested APIs (Flask/Django) and ETL processes, I bring a wealth of knowledge to the table. Dedicated to delivering clean code as evidenced by my contributions on GitHub. Capable of fixing bugs in your complex codebase as well as building a toolchain from scratch for writing and shipping code to production through fast and efficient cycles. I am currently looking for fully-remote Python engineering roles in EMEA (with guaranteed overlapping hours) or APAC time zones. EDUCATION & CERTIFICATIONS Dec 2022 Certified Kubernetes Administrator (CKA) Oct 2022 AWS Certified Solutions Architect - Associate Sep 2015 – Jun 2018 Master in Financial Markets – Neoma BS (scored 85%) Feb 2017 – Jun 2017 Exchange program in Banking & Finance – Griffith University (high distinctions) Sep 2013 – Jun 2015 Bachelor of Economics & Management – University of Brest Sep 2011 – Jun 2013 Higher National Certification in Accounting (valedictorian) WORK EXPERIENCE Jan 2023 – Jul 2023 \\ Career Break Embarked on a transformative journey, exploring 10 countries across APAC and the Caribbean— Something that has been on my bucket-list since I was a teenager. Then settling back into career. Apr 2021 – Dec 2022 \\ Quantitative Python Developer – Shell Energy Europe Ltd (London) Python | APIs | Microsoft Azure | Databricks | Data Engineering | Spark | Github | ML Ops | Terraform (IaC) Power Trading Desk Analytics: Optimizing revenue through reducing power imbalances. Machine Learning: - Forecasting solar PV and wind turbines energy production based on sensors and third-party weather data. - Statistical analysis, Seasonal ARIMAX, Recurrent Neural Networks (LSTM), STL decomposition. - Productionized models from Jupyter notebooks with ML Ops. Software Engineering (Mentored a team of 4 engineers): Modelling of UK/European markets power price forward curves taking low granularity marks and shaping half-hourly forward curves. - Refactored SDLC from clunky Excel files to battle-tested modular Python code achieving a 100% increase in test coverage (unit/regression/integration tests) and uncovering bugs in the existing codebase. - Implemented end-to-end ELT pipelines, data sourced from REST APIs (JSON payload) and transformed in Azure Databricks (Python) - Ongoing collaboration with the rest of the team (code reviews, coordinating development efforts, discussing solutions) - Successfully built and distributed Python package that contained the core business-logic with cloud-based GitOps architecture. - Built a GitHub Actions CI/CD pipeline to improve the development workflow an productionize the code. - Responsible for deploying and monitoring business critical application and troubleshooted or escalate appropriately when issues arose. - Improved deployment strategy and release management with Gitops: better rollback strategy, testing framework (Pytest). Successfully evangelized DevOps principles across the team. Python libraries: numpy, pandas, pyastral, pyspark, statsmodels, keras, scipy Jul 2019 – Mar 2021 \\ Software Engineer (Quant Research)– Prism FP (London) Python | Airflow | PostgreSQL | AWS | APIs | Kubernetes | Docker | Linux | VMs (Vagrant, VirtualBox) | OAuth2 | Helm | Terraform (IaC) | Observability (Opentracing, Grafana) Contributed significantly to the advancement of the trading analytics platform (options strategies, interest rates derivatives) by focusing on scalability, robustness, and feature expansion. Implemented test coverage and introduced new functionalities to enhance overall performance. API Development: - Continuous deployment (Jenkins, Gitlab CI) of high quality Dockerized Python code, linted (flake8, piprot) and tested (unit/integration) on a cloud-native development platform - Implemented an extensive option pricing library (volatility and rates modelling) in Python, stress tests scenarios (PCA) and other key features for the trading analytics platform (SaaS) - Microservices architecture using Flask (REST API) and instrumented our distributed system (with Opentracing and Jaeger) to monitor the latencies and error rates of our application workflow. - Built and deployed new containerized applications with Docker / Kubernetes / Helm following industry best practices. - Successfully integrated SQS queues with Lambda workers to efficiently handle infrequent workloads, troubleshooting bottleneck and ensuring our APIs remained responsive. Data Engineering: - Implemented end-to-end ELT pipelines utilizing Apache Airflow. Data extraction from diverse sources including S3, web scraping, and various vendor APIs. - Integrated Airflow on Kubernetes to enhance data pipelines scalability. - Diagnosed and resolved production issues related to data. Python libraries: numpy, pandas, flake8, PyPI server, unittests, mock, SQL Alchemy ORM (database agnostic), flask, QuantLib, asyncio Aug 2018 – Jun 2019 \\ Data Analyst – Capital Markets – Lyxor ETF (London) Python | Excel VBA | Tableau | APIs | SQL | Data Engineering Lyxor ETF was the 2nd European Exchange Traded Funds provider and second in terms of market liquidity. - Data pipelines owner (end to end) - Pre-trade analytics / Spread analysis / Market microstructure (tick data) / Market impact / Stock-exchange mecanisms - Database administration, SQL Performance Tuning / Ability to handle intensive computation with Python, vectorization of loops with numpy (slicing, masking) - Developed a Bid/Ask premium arbitrage tool to identify over/under-priced ETFs. Defining methodologies and computing metrics from tick data - Developed a web-app from scratch with Flask (blueprint, MVC, logging design patterns, front+back end) - Maintaining and improving the ETF Spreads monitoring tool (real-time data) - ETL processes in python (OO programming), Advanced SQL queries, VBA, Tableau, BQUANT, Batch - UML diagrams / Version control (Git) / Agile-Rad SDLC Jul 2017 – Dec 2017 \\ Front Office Data Analyst Intern – Lyxor ETF (Paris) Python | Excel VBA | Tableau | SQL | Data Engineering - Exposure to ETF Launches / New listings, replication strategies, performance analysis Assisting Fund Managers and Sales with Data Analysis in Python (pandas) and VBA developments Big Data analysis and visualization with Tableau SQL DB management Developed python algorithms for data quality monitoring / Web scraping / Web services (Flask) Jan 2016 – Dec 2016 \\ Cross-Asset Trader Intern – CBP Quilvest (Luxembourg) - Managing FX and cash positions - Trading FX and money market instruments, equities, bonds, options, futures, funds, ETFs, Gold - Dealing orders from direct access clients - Troubleshooting issues with Back Office and Middle Officers - Back testing performance of structured products with VBA Apr 2013 – Apr 2015 \\ Co-Founder & Treasurer – Calisthenics Sport Club Unlimiteam OPEN SOURCE PROJECTS financial-data-api (~200 hours worked) is an enterprise-grade demo project for cloud-based dockerized Flask applications(REST API) integrated with a CICD pipeline. This simplified API exposes GET endpoints that allow you to pull stock prices. In this fully-documented repo I cover the following areas: Application code : - Github Actions CICD: - static analysis: flake8, pydocstyle - Image misconfiguration/vulnerabilities (Trivy), passing artifacts between jobs - Testing patterns with Pytests (unit / integration) - Docker image multi-stage build and distribution pattern - Docker PostgreSQL DB + AWS Localstack setup for local testing - Services configuration with Docker Compose - Makefile template - Flask blueprints - Flask-SQLAlchemy implementation - Nginx (reverse proxy) and Gunicorn (WSGI) implementation - Dependency injection Infrastructure code (Multi AZ serverless architecture) : - AWS Organizations (multi-account strategy for dev & prod) with SCPs - VPC, Security-groups, RDS DB, S3, Route53, ALB, API Gateway, AWS Private link - IAM configuration (RBAC) - AWS Secrets Manager - ECS with Fargate (Blue/Green deployment) - Github Actions CICD: - Security scanner (tfsec) - Static analysis to enforce best practices (tflint, validate, fmt) - Automated infrastructure cost estimation (with Infracost) - Terragrunt patterns to keep the code DRY across environments - Automated architecture diagrams from Terraform code - Terraform remote backend bootstrap demo-elt-stack (~30 hours worked) This is a demo project (self-contained with docker-compose) that showcases an ELT process to load data into a warehouse, and process from raw layers into semantic and analytic layers with DBT. The data is then presented in a Superset dashboard. - Orchestration (Apache Airflow) Ingestion (Python) Transformation/Tests (DBT) Data visualization (Superset) Deployment (Docker-compose) kind-e2e (~150 hours worked) is a demo distributed web app running on a k8s in Docker local cluster. Its aim is to showcase fast and efficient debugging with the right tooling. Here is the tech stack that I have implemented: - Service Mesh (Istio, Kiali, Virtual Service, Gateway) - Release management (Helm charts) - Front-end: NodeJS (Express web framework) - Business-logic: Python (Django web-framework) - Observability: - Performance metrics (Prometheus) - Logs (Promtail, Loki, Grafana) - Tracing: App integrated instrumentation (Open-Telemetry, Jaeger) - Correlation between logs/traces/metrics in Grafana for faster troubleshooting developer-workstation is a Github repo in which I intend to document the tech stack I currently use. It allows me to get up to speed quicker whenever I need to setup a new workstation or when I need to transfer some knowledge to newcomers at my company for example. LANGUAGES English – Fluent French - Native speaker HOBBIES & INTERESTS Calisthenics, Thai boxing, Guitar, Natural Resources Investing, Cryptocurrencies.