diff options
Diffstat (limited to 'projects/index.qmd')
| -rw-r--r-- | projects/index.qmd | 12 |
1 files changed, 7 insertions, 5 deletions
diff --git a/projects/index.qmd b/projects/index.qmd index e7c9113..124cf02 100644 --- a/projects/index.qmd +++ b/projects/index.qmd @@ -1,7 +1,8 @@ -PROJECTS -================================================== +--- +title: "Jayrup Nakawala | Projects" +--- -[1] DECEPTIVE ALIGNMENT IN AUTONOMOUS LLM AGENTS +## [1] DECEPTIVE ALIGNMENT IN AUTONOMOUS LLM AGENTS \ Undergraduate Dissertation (Final Year) Built an agentic LLM framework to study hidden goal @@ -11,6 +12,7 @@ with constrained tools and full behavioural logging. Experiments vary perceived oversight levels and analyse deceptive tactics using: + - deterministic text filters - black-box LLM judges (user perception) - glass-box reasoning vs output analysis @@ -20,7 +22,7 @@ of hidden goals. -------------------------------------------------- -[2] SELF-HOSTED INFRASTRUCTURE +## [2] SELF-HOSTED INFRASTRUCTURE Personal infrastructure spanning home server and VPS, using Tailscale for private access and Docker Compose @@ -31,7 +33,7 @@ with minimal maintenance and reproducible setup. -------------------------------------------------- -[3] THREE REWARDS SCRAPER +## [3] THREE REWARDS SCRAPER Found the internal API used by the Three (sim company) rewards website and automated collecting rewards using |