Agent security and authorisation
Least-privilege tool access, agents bound to enterprise identity, audit trails, and threat models for prompt injection, tool poisoning and the agent supply chain.
Stacktician Limited
AI agents that are safe to let near real systems.
Architecture, security and reliability for tool-using AI agents in regulated businesses: banks, payments, insurers and anyone whose agents touch production.
Get in touchWhat Stacktician does
Production reliability
Durable execution for long-running agents: retries, idempotency, human approval steps and waits that survive a crash. Evals, tracing and token budgets so the system stays measurable and affordable.
Senior integration engineering
MCP servers and gateways, agent orchestration, cloud, CI/CD and observability, built by someone who has kept backend and payments systems running in production for twenty years.
Who
Stacktician is the practice of Arun Parihar, a software architect with over twenty years in backend, cloud and distributed systems, including payments. The current focus is the governance layer around coding agents: an issue-to-PR agent driven by durable workflows, calling tools through an MCP gateway under enterprise-issued identity, with evals and traces as the evidence. Reference implementations and write-ups are published on GitHub as they are ready.
Contact
Email hello@stacktician.com. UK-based, working remotely and on site.