Backend/Platform engineer in Minnesota with special interests in Identity/AuthX, Observability, and Messaging. Expertise in C# and F#, competent in Rust (and continually improving), interested in functional programming. M.Sc. and ten years experience in Control Systems Engineering, before transitioning to Software Engineering over a decade ago.
Available for consulting and contracting. Expertise in refactoring/replatforming large or complex legacy systems, migrating from in-house identity/auth systems to SaaS/self-hosted identity platforms, etc. I have extensive experience successfully managing/implementing complex 6-12 month projects on large business-critical production systems transparently without introducing downtime or customer impact.
Twitter: @evntdrvn
Email: eric+hn AT evntdrvn.com
Mastodon: @[email protected]
[ my public key: https://keybase.io/erics; my proof: https://keybase.io/erics/sigs/TrWi8GM4oEOpxCK3ryal0MIVE9fFCSFHjdvKMwGWJ8I ]
One thing that I learned when doing raw API LLM usage is how drastically the results can vary call per call with exactly the same input. I think that on average, people using agents underestimate the variation in results from a given turn command are, and so overindex on "X technique worked well" or "if I do Y then this will happen" or even "it did Z task well last time so it will this time too" or "{Model} is great at {thing}"
if it helps, I've found that using context (Claude.md etc) is way less effective for this type of pattern compared to using PreToolHook to capture "bad patterns" and either transparently rewriting them to "do the right thing" if that is possible statically, or if not then rejecting the tool use with a message that tells the agent "how" to use the intended tooling itself.
There was a really great article or blog post published in the last few months about the author's very personal experience whose gist was "People complain that I sound/write like an LLM, but it's actually the inverse because I grew up in X where people are taught formal English to sound educated/western, and those areas are now heavily used for LLM training."
I wish I could find it again, if someone else knows the link please post it!
I think that the messaging around this is going to be pretty important in heading off gut-reaction "it's all or nothing locked in to their world" first takes. It's probably attractive marketing for things to be aimed at "look how easy it easy to use our entire ecosystem", but there's a risk to that too.
An interesting thought experiment would be a language/toolchain that would be permissive when generating debug builds, but hard-required warn-free to generate an optimized executable.
Available for consulting and contracting. Expertise in refactoring/replatforming large or complex legacy systems, migrating from in-house identity/auth systems to SaaS/self-hosted identity platforms, etc. I have extensive experience successfully managing/implementing complex 6-12 month projects on large business-critical production systems transparently without introducing downtime or customer impact.
Twitter: @evntdrvn Email: eric+hn AT evntdrvn.com Mastodon: @[email protected]
[ my public key: https://keybase.io/erics; my proof: https://keybase.io/erics/sigs/TrWi8GM4oEOpxCK3ryal0MIVE9fFCSFHjdvKMwGWJ8I ]