AI Engineer / Agentic Systems / Retrieval & Memory / Full-Stack Products
I build the version that survives contact.
AI Systems Engineer @ RYSE
Somay Kousis / Co-Founder, Something
I build stateful multi-agent systems, retrieval and memory pipelines, and product-grade software. The work I care about is the part that holds up under contention, cost limits, and being wrong, not the demo that runs once.
Built the orchestrator routing tasks across a 3,000+ subagent registry through intent classification, capability scoring, context assembly, and dispatch, cutting misrouted tasks ~40% and dispatch latency to ~120ms.
02 / Proof
Kept 25 of 25 concurrent writes on a contended counter, with 58 SERIALIZABLE conflicts caught and auto-retried, where a flat-file analog kept 1 and silently lost 24. Ships as a runnable benchmark, not a screenshot.
03 / Proof
Cut CI log payload 76% to 93% before any model call, measured against Podman's own logformatter fixtures and reproducible from a fresh clone with no token and no API budget.
Experience
Shipping in production, not in slides.
Orchestration at scale, a company I co-founded, and client work delivered end to end. Each one is listed with the part that was actually hard.
01 / Remote
RYSE Technologies Pvt Ltd
AI Systems Engineer Intern
Jul 2026 – Sep 2026
Building the orchestration layer that routes work across a large registry of specialized subagents.
/
Engineered 5+ production-grade subagents using UV and LangGraph, implementing persistent memory layers for long-running task continuity.
/
Architected a central orchestrator agent to route 10,000+ daily tasks across specialized subagents, reducing misrouted tasks by 33% and dispatch latency by 24%.
Call a real race condition an infra blip and you have told a maintainer to press re-run on a genuine bug. So abstention became a first-class verdict, and the harness reports abstention rate next to accuracy and counts the dangerous confusion separately. A classifier that says nothing is recoverable. One that is confidently wrong is not.
From Podman Flake Agent2026.07
Prompts Are Requests, Schemas Are Constraints
Asked politely for two perspectives, the graph produced six to eight, a 3 to 4x overrun that burned the token ceiling before the debate resolved. No amount of prompt rewording fixed it. Moving perspective count into a state schema the moderator schedules against fixed it completely, because the constraint stopped being something the model could ignore.
From RabbitHole2026.07
Never Overwrite What You Might Have To Explain
A fact is not static. It gets recorded, superseded, contradicted, and sometimes restored. Destructive updates make the current state cheap to read and the history impossible to recover. Closing rows with valid_to instead of overwriting them keeps a harder question answerable: what did we believe last month, and were we right?
From PaperPlanes
The short version
Start with the projects. The proof is in the repos.
I am open to AI engineering work where agentic workflows, retrieval, and product implementation meet. Every claim on this site traces to a case study and a repository, so the fastest way to evaluate me is to read one and then go check it.
