Services
What we build
Four kinds of work, each sold as a fixed-scope sprint or a monthly retainer.
AI Voice & Chat Agents
Agents that answer the phone, qualify leads, and hand off to a human.
Production voice and chat agents for inbound intake, triage, scheduling, and follow-up. We pick the voice model per use case, for latency, voice quality, and cost, and wire it into your phone system, with warm transfer to your team and evals that test call behavior before each deploy.
What you get
- Inbound or outbound agent for a defined use case
- Phone numbers, SIP, and warm transfer to your staff
- Structured call summaries and leads pushed to your systems
- Eval suite that regression-tests calls before each deploy
- Call logs and monitoring
Best fit: Law firms, medical practices, and service businesses losing callers after hours or paying staff to handle repetitive intake.
AI Features in Your Product
LLM integrations, MCP servers, and evals for the product you already have.
We add AI to existing products and workflows: document processing, drafting, search, classification, and agents that take actions. That includes MCP servers and OAuth so your customers can use your product from Claude and ChatGPT, plus the evals that tell you whether a prompt or model change made things better or worse.
What you get
- LLM integration (Claude, OpenAI, Gemini) in your stack
- MCP server and OAuth 2.1 for Claude and ChatGPT access
- Eval suite and regression tests for AI behavior
- Monitoring, cost tracking, and a runbook for your team
Best fit: Software teams and SMBs with a working product who want AI features that hold up in production.
Full-Stack Web & iOS Apps
Web apps, internal tools, and native iOS apps, from idea to production.
Full-stack builds in Next.js and TypeScript, native iOS in Swift and SwiftUI, and the APIs, databases, and background jobs behind them. BookMatcher, Recall Beacon, and the AOTL field-operations app are representative builds.
What you get
- Discovery and scope
- Next.js + TypeScript web app and/or native SwiftUI iOS app
- API, database, auth, payments, and integrations
- Deployment, App Store submission, and handoff
Best fit: Founders launching a product and businesses replacing spreadsheets, paper, or manual workflows.
Codebase Rescue & Hardening
Security fixes, CI/CD, and Terraform for code you inherited.
For products that are stuck: launch blockers, fragile deploys, or security gaps nobody has had time to close. We find and fix the access-control holes first, then move infrastructure to Terraform, previewed on every pull request, with automated deploys and alerting. A medical case-management platform is the reference engagement.
What you get
- Security review and fixes for access control and data exposure
- Critical bug and launch-blocker fixes
- Terraform infrastructure with plans on every PR
- Automated CI/CD, monitoring, and alerts
Best fit: Founders and operators who inherited a codebase, or whose last developer left, and need it stable and secure.
How we work
Fixed-scope sprint
A defined outcome, scope, and timeline, agreed before work starts. Best for a first build, a voice agent, or a rescue.
Monthly retainer
Ongoing engineering capacity for teams that need a senior builder on call: new features, AI work, and on-call support.
AI-native process
Agentic coding, test-first changes, and human-reviewed pull requests. You get speed without giving up review or ownership of the code.
Not sure which fits?
Tell us what you’re trying to accomplish. We’ll point you to the right starting point on a short fit call.
Book a fit call