Senior Software Engineer
George
Feighan
Architecting and building across the full stack — from data platforms and cloud infrastructure to AI tooling and microservices. Strong on technical execution; equally comfortable with stakeholders, requirements, and the "why" behind what gets built.
Full-Stack Engineer with a Data & AI Focus
I'm a Senior Software Engineer with a broad skill set spanning backend development, data engineering, cloud infrastructure, and AI integration. I'm comfortable working across the full stack — picking up whatever the problem requires rather than staying in one lane — and I have a track record of taking ownership of complex, ambiguous work and seeing it through end to end.
On the data and AI side, I've designed and built production data platforms, implemented LLM-powered tooling across multiple model providers, and worked with modern data stack technologies including PySpark, dbt, and Apache Iceberg. On the backend side, I have deep experience with .NET/C# and Python across microservices, REST APIs, and cloud-native architectures on AWS and Azure. My background in computational physics shapes how I think about systems — analytically, and from first principles.
I also care about the product and people side of engineering — translating business requirements into technical decisions, working directly with stakeholders, and communicating clearly across technical and non-technical audiences. I've mentored junior developers and led development end-to-end, from requirements gathering through to delivery and user training.
I've worked across a range of company sizes and industries — from Big 4 professional services and large payment providers through to early-stage startups where I was the primary engineer. That breadth spans fintech, utilities, pharmaceuticals, banking, and professional services, and has given me a practical understanding of how engineering looks different depending on the environment you're operating in.
Technical Stack
Career History
- Sole architect and engineer of the company's data lakehouse platform — AWS Glue (PySpark) ETL ingesting from PostgreSQL, HubSpot, Stripe, Intercom, and SendGrid into Apache Iceberg on S3.
- Implemented dbt transformations on ECS Fargate and auto-deployed Power BI semantic models via XMLA.
- Built an AI-powered Slack investigator bot using AWS Bedrock (Claude) with parallel Athena query execution, enabling natural language data querying across the team.
- Developed an AI-driven energy quote automation tool using Playwright MCP browser agents, a multi-model LLM abstraction layer (Claude, GPT-4, Gemini, Bedrock), and GCP Cloud Run.
- Delivered major partner integrations end-to-end, collaborating directly with stakeholders to design bespoke solutions.
- Primary backend engineer for the core platform — supporting .NET/C# microservices and a shared NuGet library.
- Provisioned all AWS infrastructure with Terraform across dev/UAT/prod including Lambda, ECS, and Datadog observability.
- Rapidly onboarded into a large, complex payment services codebase and began delivering immediately alongside cloud architects, infrastructure engineers, and testers.
- Migrated a platform from Azure Service Fabric to containerised Web APIs and extended email service functionality using C#.NET, Azure Service Bus, APIM, and Redis Cache.
- Wrote unit tests with NUnit throughout to maintain coverage on delivered changes.
- Primary engineer responsible for all areas of technology, working across the full stack from backend APIs to frontend and infrastructure.
- Built the platform using ASP.NET Core, ReactJS, and a microservice architecture on Azure — with Azure Functions, Cosmos DB, and Azure AD B2C for authentication including custom policy flows.
- Implemented RAG AI search using OpenAI and vector indexes, enabling users to ask context-specific questions about their content, autocomplete text, and generate new material.
- Implemented SEO best practices and tracked search and user analytics via Google Analytics and MS Clarity, with PowerBI dashboards for reporting.
- Set up CI/CD pipelines via Azure DevOps and wrote comprehensive unit and integration tests with XUnit.
- Hired as the first developer in the Valuations practice, working independently alongside senior accountants to build the team's technology capability from scratch.
- Built financial analysis platforms using ASP.NET MVC, React, Entity Framework, and SQL on Azure — covering discount rates, trading multiples, bond rates, and volatility scores — saving thousands of client-facing hours and generating over £1m in annual revenue.
- Owned the full development lifecycle: requirements gathering with the business, development, deployments, and delivering training sessions and presentations to large audiences.
- Worked in a technical BA capacity on additional projects — including using Gazette data to support business restructuring teams and providing technical guidance on a carbon valuation project.
- Firm-wide SME for the S&P Capital IQ API across the Deals line of service; career coach to a junior developer at Manager grade.
- Consistent delivery helped grow the Valuations technology team from 3 to over 20 people with significantly expanded business funding.
- .NET developer at a consultancy, completing long-term placements at a global pharmaceutical company (database migration) and a major bank (banking support tool).
Technical Expertise
A broad stack built across a near-decade of engineering — from enterprise .NET development at PwC to data engineering and AI tooling at Bunch.
Languages
Backend & Web
Cloud & Infrastructure
Data Engineering
AI & LLM Integration
Product & Practices
Notable Work
A selection of projects that reflect the range of problems I've worked on.
Data Lakehouse & AI Querying
Designed and built Bunch's data lakehouse from scratch — AWS Glue PySpark ETL jobs ingesting from PostgreSQL, HubSpot, Stripe, Intercom, and SendGrid into Apache Iceberg on S3, with dbt transformations on ECS Fargate and Power BI semantic models auto-deployed via XMLA. Built on top of this, an AI-powered Slack bot (AWS Bedrock / Claude) lets the whole team query the data in plain English via parallel Athena queries — no SQL required.
Energy Quote Automation
Developed an AI-driven tool that automates the energy quote process using Playwright MCP browser agents. Built a multi-model LLM abstraction layer supporting Claude, GPT-4, Gemini, and AWS Bedrock — deployed on GCP Cloud Run.
RAG AI Search
Implemented context-aware AI search across a note-taking platform using OpenAI and vector indexes. Users could ask questions specific to their own content, autocomplete sentences, and generate new material — all grounded in their personal data rather than general LLM knowledge. Built as part of a broader full-stack platform on ASP.NET Core, ReactJS, and Azure.
PwC Valuations Platform
Built financial analysis tools for PwC's Valuations practice, from scratch as the first developer hired. Platforms handled complex calculations — discount rates, market multiples, volatility scoring — and generated over £1m in annual revenue. Helped grow the engineering team from 3 to over 20.
Get in Touch
Feel free to reach out via email or connect on LinkedIn.