Production-Grade Software & Private AI Systems
Tannahill Systems is an independent software engineering studio and research lab. We architect custom full-stack web platforms, local LLM inference pipelines, distributed data scrapers, and native cross-platform tooling for businesses and technical teams.
End-to-end delivery across Next.js, TypeScript, .NET 9, FastAPI, and PostgreSQL with strict type safety.
Self-hosted LLM orchestration, custom RAG indexing, and QLoRA fine-tuning without third-party data exposure.
Clean, maintainable repository handoffs with containerized deployments and zero vendor lock-in.
Custom Software Built for Production Scale
We partner with founders, enterprises, and operational teams to design, build, and deploy mission-critical software systems from greenfield concept to cloud release.
Full-Stack Application Engineering
End-to-end architecture and implementation of scalable web platforms, customer-facing SaaS products, and high-concurrency API backends.
Core Deliverables
- Multi-tenant SaaS applications & custom web portals
- RESTful & real-time WebSocket API microservices
- Relational database schema design, indexing & migrations
- Authentication, RBAC & automated CI/CD deployment pipelines
Private AI & Local LLM Infrastructure
Design and deployment of custom Retrieval-Augmented Generation (RAG) pipelines, quantized local model inference servers, and domain-specific fine-tuning workflows.
Core Deliverables
- Private on-premise or VPC-hosted LLM inference APIs
- Hybrid semantic + vector search RAG architectures
- QLoRA model fine-tuning & synthetic dataset generation
- Automated LLM evaluation harnesses & latency benchmarking
Data Extraction & Cloud Automation
Resilient distributed scraping engines, automated ETL ingestion pipelines, and serverless cloud workflows engineered for real-time market intelligence.
Core Deliverables
- Headless browser & API data extraction pipelines
- Automated data normalization, deduplication & warehousing
- Event-driven serverless compute & scheduled background workers
- Real-time analytics dashboards & alerting webhooks
Native Desktop & Developer Tooling
Cross-platform desktop applications, CLI automation utilities, and IDE extensions built for minimal memory overhead and native execution speed.
Core Deliverables
- Cross-platform native desktop apps (Windows, Linux, macOS)
- Custom VS Code extensions & language server integrations
- Static codebase analysis scanners & AST transformation tools
- Open-source SDKs and published package distribution
Platforms Engineered In-House
Alongside client engagements, Tannahill Systems actively develops and maintains proprietary AI runtimes, static analysis engines, and cloud data platforms.
LocalChat Inference Engine
Enterprise .NET 9 Local LLM Runtime & Memory Orchestrator
A modular C# and .NET 9 Web API inference engine providing structured context retrieval, sliding-window memory management, and automated conversation summarization for local large language models.
CodeCurator & MDV
High-Performance Codebase Analysis & Native Desktop Tooling
Extensible static analysis orchestrator with plugin-based scanner execution paired with MDV, a lightweight cross-platform desktop Markdown rendering application built on Rust and Tauri.
LegisTrack & TrendDev
Automated Legislative & Market Intelligence Platforms
Full-stack data ingestion and AI summarization platforms that track United States legislative updates and analyze software engineering market demand across serverless cloud infrastructure.
compel5 & py-data-gen
Prompt Weighting SDK & Synthetic Dataset Pipelines
Published PyPI library for embedding conditioning and prompt weighting in diffusion and transformer pipelines, paired with automated synthetic dataset generators for QLoRA fine-tuning.
Core Infrastructure & Delivery Workflow
Every system is built on battle-tested, modern runtimes designed for low operational overhead, high throughput, and seamless horizontal scaling.
Languages & Runtimes
- TypeScript
- C# (.NET 9)
- Python
- Rust
- SQL
- Bash / PowerShell
Frontend & Desktop
- Next.js 15
- React 19
- Tailwind CSS
- Tauri
- shadcn/ui
- HTML5 / CSS3
Backend & Data
- ASP.NET Core
- FastAPI
- PostgreSQL
- Supabase
- AWS DynamoDB
- Redis
AI & Cloud Infrastructure
- llama.cpp / Ollama
- vLLM
- Unsloth / QLoRA
- AWS (S3, Lambda)
- Docker
- Cloudflare Edge
Client Engagement Workflow
Technical Discovery & Architecture Specification
We audit your operational requirements, define data models, select the optimal runtime stack, and deliver a concrete technical specification with fixed milestones.
Modular Iterative Engineering
Software is constructed in decoupled, type-safe modules with continuous staging deployments, automated testing, and transparent repository access.
Production Deployment & Telemetry
We provision production cloud or on-premise infrastructure, configure SSL/DNS security, verify load performance, and execute full documentation handoff.
Initiate an Engineering Engagement
Whether you need a custom full-stack web platform, a private RAG/LLM deployment, or automated data extraction infrastructure, reach out with your technical objectives.
Operations Base
Illinois, United States
Serving clients nationwide (US Central Time)