Building production AI systems where latency and correctness aren't negotiable —
from sub-100ms derivatives trading to sub-600ms real-time voice AI.
- Fintech — Low-latency trading and regulated financial systems: sub-100ms derivatives pipelines (~35k trades/day) and regulatory intelligence over 1M+ documents.
- AI/ML — Production LLM, RAG, multi-agent and real-time voice AI (p95 < 600 ms), with fine-tuning and latency-tiered serving.
- Distributed systems — Event-driven architectures on Kafka, real-time processing in Go and Python.
- Cloud & reliability — AWS/EKS and Kubernetes with OpenTelemetry and SLI/SLO; 99.95% platform uptime.
Languages
AI / LLM
Backend & Web
Cloud & DevOps
Data



