Rehan Arvi (rehanarvi1605)

Structured Multi-Step Agent Brittleness & Evaluation Harness
Structured Multi-Step Agent Brittleness & Evaluation Harness

A structured multi-step LLM workflow that evaluates the reliability of multi-step agents, surfaces concrete failure modes, and produces actionable hardening recommendations far beyond a single prompt.

0
KinLore Journal
KinLore Journal

A Family legacy journal that prioritize Privacy that passively extracts longitudinal cognitive biomarkers leveraging deterministic math and secure LLM inference

0
KinLore
KinLore

A Family legacy journal that prioritize Privacy that passively extracts longitudinal cognitive biomarkers leveraging deterministic math and secure LLM inference.

0
Hyper_Mapper
Hyper_Mapper

Hyper-Mapper: Adapting rigid curricula to the neurodivergent brain through cognitive translation.

0
LLM Ledger — AI API Cost Intelligence
LLM Ledger — AI API Cost Intelligence

See exactly where your AI budget is going. By model. By feature. By user.

0
ContextOps: Orbit-Powered Engineering Intelligence
ContextOps: Orbit-Powered Engineering Intelligence

Stop guessing ! ContextOps uses GitLab Orbit's SDLC graph to give AI agents true context,instantly mapping issue complexity and detecting pipeline drift across your entire org.

0
Time_Bank
Time_Bank

Don't limit screen time,earn it !!. Time Bank locks distracting apps until you complete physical tasks verified by AI. Trade sweat for dopamine, or face extreme mental friction.

0
CodeDNA
CodeDNA

Code DNA audits model "genetics" in real-time, catching adversarial prompts and clinical hallucinations by analyzing internal activations before they become output.

0
Overwatch Triage
Overwatch Triage

Overwatch Triage aggregates messy Reddit queues into visual text clusters using Redis, deploying asynchronous background workers to safely bulk-delete target raids without platform timeout crashes.

0
ClaimGate Enterprise
ClaimGate Enterprise

Tactical AI infrastructure utilizing a multi-gate interceptor process to secure enterprise LLM deployments against Out-of-Domain attacks.

0
ProcureGuard AI
ProcureGuard AI

ProcureGuard AI: 100% Math. 0% Hallucinations.

0
Vigilens-Cognitive-Assistant
Vigilens-Cognitive-Assistant

Vigilens: Moving medical AI out of the text box and into the real world

0