Portfolio

#!/bin/bash

echo "You either die a hero... or live long enough to become the villain."
echo "I chose to code."

a bit about myself

• building at the edge of ai + product, where infra meets ideas, and latency hits real users.
• right now working as a gen ai dev @kliqstr — built 5+ LLM agents, a custom rag setup (boosted accuracy by 40%), and scaled stuff with aws ecs + lb for 100+ users. [Feb’24 - Aug’24]
• before that, fullstack @occultdiy handled backend + infra, docker, redis, hit 100k+ req/day. [Sept’24 - Aug’24]
• also worked at @amd telecom — built invoice system (95% acc) using llms + setup image gen pipeline with stable diffusion + controlnet. [Mar’23 - Mar’24]
• just trying to build useful, fast, and a bit fun tools tbh.

my interests

• kinda obsessed with systems that scale. like digging into bottlenecks, caching, infra stuff feels like puzzle solving.
• love building agents, rag flows, llms that actually do stuff, not just reply.
• mostly in backend docker, redis, celery, queues... that messy stuff feels peaceful.
• also making side tools where ai writes, edits, builds apps.
• i don’t touch frontend much, but yeah i care about clean ui too.
• and yeah… lowkey into geopolitics and tech policy. feels like debugging, but global.

what i’m working on?

• shipping llm agents to prod tuning, rag stuff, trying to make them faster + useful.
• playing around with infra docker, ecs, load balancer, queues... just trying to keep it stable and cheap 😁.