Forward-deployed AI engineer · Production voice & LLM systems · Python, Node, GCP
-
KyrosAura Corp Pvt Ltd
- Mumbai, Maharashtra
-
06:56
(UTC -12:00) - https://kyrosaura.com/
- in/saurabh-rajvanshi
Pinned Loading
-
LiveKit onboarding: seven findings f...
LiveKit onboarding: seven findings from a first-time Windows install 1# LiveKit: notes from a first-time onboarding23**Saurabh Rajvanshi** · Windows 11 · LiveKit CLI 2.17.0 · agents 1.6.5 · `agent-starter-python`
414 July 2026 · clean machine, no prior LiveKit experience · two console sessions, ~10 minutes of speech5 -
ctaio-reads
ctaio-reads PublicRubric-based LLM evaluation pipeline. Model scores and drafts; deterministic code owns validation, filtering, and publication. Fail-closed — rejections logged, never repaired.
JavaScript
-
LLM-evaluation-problem-Harmonic-Symphony-Partitions
LLM-evaluation-problem-Harmonic-Symphony-Partitions PublicCompetitive programming problem designed to expose failure modes in frontier LLMs. Features a 3-layer trap invalidating greedy, O(N²) DP, and prefix-sum optimizations. Correct O(N log V) solution v…
Python
-
sg-engineer-004
sg-engineer-004 PublicProduction-grade system design challenge for a real-time analytics platform. Engineering design, architecture decisions, operating artifacts, and supporting documentation.
Python
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.