Building Ask Warren: A Production RAG System with 47 Years of Investment Wisdom
How I built a retrieval-augmented generation system that lets you chat with Warren Buffett's shareholder letters - to understand RAG trade-offs in production.
bullet-proof backends
fast
I'm Steven Mays, a Software Architect building scalable, serverless solutions and AI-driven platforms that reduce costs, accelerate growth, and drive measurable business outcomes.
About Me
Impact & Results
Measurable outcomes that demonstrate the power of strategic technical architecture.
Migrated legacy infrastructure to serverless architecture, dramatically reducing operational overhead.
Built automated ad deployment system handling $100K+ daily budgets across Facebook, Google, and TikTok, replacing a 10-person team.
Optimized database queries and API responses for high-traffic production systems.
Scaled consulting firm from solo founder to 12-person team in under 2 years.
Technical Proficiency
Core Expertise
Delivering real-world solutions that transform complex business problems into measurable successes.
Architecting event-driven microservices, APIs, and real-time data processing systems on AWS, reducing operational overhead and costs.
Building production-grade AI tools like the startup generator, leveraging Google Gemini API with robust JSON parsing and cost-effective serverless architecture.
Elevating engineering teams through hands-on mentorship, code reviews, and effective knowledge-sharing programs.
Technical Stack
Deep experience across modern development languages, frameworks, cloud infrastructure, and databases.
Node.js, TypeScript, Go, C#, GraphQL, NestJS, Serverless Framework, ASP.NET
AWS (Lambda, S3, DynamoDB, API Gateway, CloudFormation), Docker, Kubernetes, CI/CD
Facebook Marketing API, Google Ads API, TikTok Ads, CAPI, Google Tag Manager, Segment, Lookalike Audiences
PostgreSQL, MongoDB, MySQL, DynamoDB, Redis, Elasticsearch
Event-Driven Architecture, NATS, RabbitMQ, AWS SQS/SNS, Kinesis
Generative AI Integration, LLM Prompting, Google Gemini, Claude, GPT
Industry Experience
Automated ad deployment system managing thousands of campaigns with server-side tracking, affiliate portals, and custom attribution systems generating $1M+ annual revenue impact.
Secure and compliant backend systems powering mobile payments, loyalty programs, and investment management.
Robust backend solutions for real-time warehouse automation, asset tracking, and enterprise integrations.
Founded a tech consultancy, scaling rapidly and providing CTO-level technical architecture and strategic guidance to early stage startups.
Need a seasoned engineer or AI strategist? Let's collaborate and turn your ideas into impactful realities.
Sharing practical advice and proven techniques from decades of engineering and development experience.
How I built a retrieval-augmented generation system that lets you chat with Warren Buffett's shareholder letters - to understand RAG trade-offs in production.
How I built a complete AI-powered web application using agentic coding - from concept to production in hours, not days.
Unlock the full potential of your coding workflow by mastering high-leverage techniques with AI.
Right now, my daughters won't remember any of these early years—their first steps, first words, the little moments we share each day. But this time is still critical. This is when their Self is formed, and each interaction is helping shape that inner voice they'll carry through life.