Speaking
Travis Colbert speaks on full-stack engineering, how to leverage AI with right-sized solutions and best practices. He has a special place in his heart for small, comprehensible systems over framework-heavy architectures.
Topics
- Building without a framework. What you gain and what you give up when the whole server is a request handler and a few modules. Drawn from Zetta, a blog engine with one dependency.
- How to build AI safely into business processes. What does it take to implement AI solutions into any sized business? How to be safe about it. What to expect.
- Tool-based retrieval for LLM applications. Why a handful of well-described tools often beats a vector database for small, bounded corpora.
- Reducing dependencies as a maintenance strategy. The long-term cost of a dependency tree, and what it takes to run without one.
Formats
- Conference talk, 25–45 minutes, with or without Q&A
- Workshop, half day, hands-on, capped at about 20 people
- Podcast or recorded interview, remote
- Internal engineering talk for a company audience
Availability
Booking roughly one quarter out. Remote events are easier to schedule than travel; happy to travel for a full talk or workshop when the timing works.
Rates
Depends on format, travel, and whether the event is commercial, community-run, or internal to a company. Community and non-profit events are often free. Ask and Travis will quote the specific event.
Travel and accommodation are expected to be covered for in-person events outside the local area.
What to include when asking
Event name, date, location or remote, audience size, talk length, and whether the event is commercial or community-run. A reservation can be made through this chat.