┌────────────────────────────┐ │ SUMMARY │ ├────────────────────────────┤ │ mov rax, "about me" ; ret │ │ │ │ self-taught since │ │ before i could legally │ │ drive. i live at the │ │ intersection of │ │ kernels, networking, │ │ firmware, and whatever │ │ rabbit hole i fell │ │ into this week. built │ │ an ISP, sold a company │ │ once. write Rust that │ │ would make most people │ │ uncomfortable. │ │ │ │ daily-drive mainline │ │ kernels by choice, not │ │ accident. every device │ │ i own has run linux │ │ for the last 18 years. │ │ run BGP in production │ │ and think about it in │ │ the shower. │ │ │ │ // i never said i was │ │ stable. the code │ │ usually is though. │ └────────────────────────────┘
╔══════════════════════════════════════╗ ║ SUMMARY ║ ╠══════════════════════════════════════╣ ║ mov rax, "about me" ; ret ║ ║ ║ ║ self-taught since before i could ║ ║ legally drive. i live at the ║ ║ intersection of kernels, ║ ║ networking, firmware, and ║ ║ whatever rabbit hole i fell into ║ ║ this week. built an ISP, sold a ║ ║ company once. write Rust that ║ ║ would make most people ║ ║ uncomfortable. ║ ║ ║ ║ daily-drive mainline kernels by ║ ║ choice, not accident. every ║ ║ device i own has run linux for ║ ║ the last 18 years. run BGP in ║ ║ production and think about it in ║ ║ the shower. ║ ║ ║ ║ // i never said i was stable. ║ ║ the code usually is though. ║ ╚══════════════════════════════════════╝
╔══════════════════════════════════════════════════════════╗ ║ SUMMARY ║ ╠══════════════════════════════════════════════════════════╣ ║ mov rax, "about me" ; ret ║ ║ ║ ║ self-taught since before i could legally drive. i ║ ║ live at the intersection of kernels, networking, ║ ║ firmware, and whatever rabbit hole i fell into this ║ ║ week. built an ISP, sold a company once. write Rust ║ ║ that would make most people uncomfortable. ║ ║ ║ ║ daily-drive mainline kernels by choice, not ║ ║ accident. every device i own has run linux for the ║ ║ last 18 years. run BGP in production and think about ║ ║ it in the shower. ║ ║ ║ ║ // i never said i was stable. the code usually is ║ ║ though. ║ ╚══════════════════════════════════════════════════════════╝
╔══════════════════════════════════════════════════════════════════════════════╗ ║ SUMMARY ║ ╠══════════════════════════════════════════════════════════════════════════════╣ ║ mov rax, "about me" ; ret ║ ║ ║ ║ self-taught since before i could legally drive. i live at the ║ ║ intersection of kernels, networking, firmware, and whatever rabbit hole ║ ║ i fell into this week. built an ISP, sold a company once. write Rust ║ ║ that would make most people uncomfortable. ║ ║ ║ ║ daily-drive mainline kernels by choice, not accident. every device i own ║ ║ has run linux for the last 18 years. run BGP in production and think ║ ║ about it in the shower. ║ ║ ║ ║ // i never said i was stable. the code usually is though. ║ ╚══════════════════════════════════════════════════════════════════════════════╝
╔══════════════════════════════════════════════════════════════════════════════╗
║ EXPERIENCE ║
╠══════════════════════════════════════════════════════════════════════════════╣
║ ~/career ║
║ │ building an ISP that actually gives a damn about the people incumbents ║
║ │ forgot. ║
║ │ ║
║ │ BGP peering, fiber infrastructure, network engineering, upstream ║
║ │ wrangling, ║
║ │ customer ops. yes the name is real. yes the ASN is live. yes the routing ║
║ │ tables are beautiful and i am very normal about them. ║
║ │ ║
║ │ // building the internet of tomorrow ║
║ │ applied AI research on the software engineering side. trained models, ║
║ │ built the infrastructure to train and measure them, and instrumented ║
║ │ existing systems to squeeze research data out of them. ║
║ │ ║
║ │ what i shipped: ║
║ │ ║
║ │ LLM fine-tuning - supervised fine-tuning of large language models for ║
║ │ structured, domain-specific tasks. curriculum-based data generation, ║
║ │ parameter-efficient training (LoRA) on single GPUs and full fine-tunes ║
║ │ distributed across multi-GPU clusters. wrote the eval harnesses to ║
║ │ actually verify the models learned what they were supposed to. ║
║ │ ║
║ │ ML training infrastructure - internal training-observability tooling: ║
║ │ thread-safe metric tracking, distributed-aware aggregation, anomaly ║
║ │ detection, low overhead by design. strict typing, high coverage, CI. ║
║ │ shipped. ║
║ │ systems instrumentation - research-grade logging bolted into existing ║
║ │ high-performance systems to quantify where the work was actually ║
║ │ happening, without touching their behavior. ║
║ │ ║
║ │ infra - self-hosted internal services and deployment automation. ║
║ │ ║
║ │ // research is just breaking things on purpose and writing down what fell ║
║ │ out. ║
║ │ frontend and backend dev for InCenter — an automated breach and attack ║
║ │ simulation platform covering web, network, and cloud security. ║
║ │ ║
║ │ deployed and configured InCenter infra on AWS using terraform and ║
║ │ ansible. ║
║ │ ║
║ │ // the job where breaking things was the actual job description. ║
║ │ network software engineering on proxy infrastructure at scale. ║
║ │ yes i thought about packets constantly. yes that was fine. ║
║ │ solo founded and ran an AI automation and marketing company. just me. ║
║ │ built everything, ║
║ │ sold everything, supported everything. ║
║ │ ║
║ │ most of my clients were adult content creators was on OnlyFans, Fansly, ║
║ │ Twitter, ║
║ │ Tumblr, Cam Sites, etc that whole world. also had mainstream social media ║
║ │ growth clients and social media marketing agencies work with me to help ║
║ │ their customers (YouTube, Instagram, Spotify, Discord, the usual). the ║
║ │ adult industry was the bulk of it. ║
║ │ ║
║ │ honestly the adult entertainment industry pays better and the technical ║
║ │ problems are more ║
║ │ interesting. ║
║ │ ║
║ │ what i shipped: ║
║ │ ║
║ │ - stable diffusion pipelines + custom AI image editing tooling, ║
║ │ you know, before "generative AI" was a buzzword anyone used ║
║ │ - conversational AI agents handling inbound DMs and fan interactions ║
║ │ ║
║ │ - full platform automation: posting, scheduling, shares, reposts, ║
║ │ likes, replies, follows running simultaneously across Twitter, ║
║ │ Instagram, Tumblr and more, hands-free ║
║ │ - automated transaction handling fan pays, gets acknowledged, ║
║ │ money routes to the creator, zero manual intervention ║
║ │ - audience targeting and growth systems that turned one-time ║
║ │ followers into recurring paying subscribers for my clients ║
║ │ - the whole stack was designed so clients checked a dashboard ║
║ │ occasionally and otherwise forgot it existed ║
║ │ ║
║ │ retention was the metric that mattered. clients stayed because their own ║
║ │ subscriber bases kept growing and paying. ║
║ │ recurring revenue on both ends. ║
║ │ ║
║ │ ran it all solo. customer acquisition, support, dev, infra, billing ║
║ │ ║
║ │ scaled it to the point someone made an offer on the whole thing: ║
║ │ software, IP, client relationships, resources, everything. ║
║ │ asset acquisition. ║
║ │ NDA covers buyer identity and terms. ║
║ │ ║
║ │ // reverse engineered platform APIs, built thin clients for major social ║
║ │ and messaging services, ran adult content automation AI infrastructure at ║
║ │ scale. from my bed. ║
║ │ "before it was cool." ║
║ │ built and compiled custom kernels. ported linux to embedded devices. ║
║ │ wrote kernel-level and socket-level code. developed and modified firewall ║
║ │ and security software. custom linux firmware with hand-applied patches. ║
║ │ ║
║ │ racked servers, NAS, switches. configured DDNS, DHCP, name servers. ║
║ │ debugged hardware issues. ran daily internal and external pentests. ║
║ │ organized red and blue teams on the company network. wrote security ║
║ │ reports and patching plans. automated scanning of new database leaks ║
║ │ for employee records. hired and trained new employees. ║
║ │ ║
║ │ // if it compiles it ships. learned why that's a bad philosophy here. ║
║ │ domain user and group management. rewrote linux and unix documentation ║
║ │ that hadn't been touched in years. ║
║ │ ║
║ │ // the documentation was basically archaeology ║
║ │ ║
║ │ resolved undocumented server issues ║
║ │ for clients and engineers. fixed ICA remote client connectivity. ║
║ │ sorted out aruba router, VOIP, and firewall configs. ║
║ │ rebooted the rdps; ║
║ │ ║
║ │ // resolved bugs that would have led to over 6 million in investments to ║
║ │ disapear in under 5 minutes ║
║ │ test servers, storage, switches and sub-assemblies. ║
║ │ verify proper components are received per order pick list prior to build. ║
║ │ scan all components into the manufacturing shop floor system to ensure ║
║ │ proper parts and inventory tracking. ║
║ │ use online methods and work instructions to build racks per customer ║
║ │ specification. ║
║ │ built internal tooling to automate analytics and invoicing. ║
║ │ found an active RCE exploit a former employee had left running. killed ║
║ │ it. ║
║ │ debugged a broken SMTP server. built webservers for employee training, ║
║ │ interview management, and onboarding. ║
║ │ ║
║ │ // did incident response at a pizza place. it counted. ║
║ │ network monitoring, security config, access permissions, diagnosing ║
║ │ networking problems with diagnostic tooling, routine maintenance. ║
║ │ ║
║ │ // early IT. everyone starts somewhere. ║
║ │ maintained schools networks, internally and externally tested and ║
║ │ reported security flaws and vulnerabilities district wide, fixed teachers ║
║ │ and student machines, set up classroom networks and devices. ║
║ │ ║
║ │ ║
║ │ // tldr: i got into 'ethical hacking' by finding holes in networks i was ║
║ │ supposed to be using to do homework on ;3 ║
║ │ // classic origin story ║
║ fixed arcade and vending machines. soldering, wiring diagrams, replacing ║
║ mechanical and electrical parts. diagnosed malfunctions, ordered parts, ║
║ handled billing. ║
║ ║
║ // this is where i learned that hardware is just software you can hit ║
╚══════════════════════════════════════════════════════════════════════════════╝
╔════════════════════════════════════════╗
║ EXPERIENCE ║
╠════════════════════════════════════════╣
║ ~/career ║
║ │ founder ║
║ │ ║
║ │ building an ISP that actually gives ║
║ │ a damn about the people incumbents ║
║ │ forgot. ║
║ │ ║
║ │ BGP peering, fiber infrastructure, ║
║ │ network engineering, upstream ║
║ │ wrangling, ║
║ │ customer ops. yes the name is real. ║
║ │ yes the ASN is live. yes the ║
║ │ routing ║
║ │ tables are beautiful and i am very ║
║ │ normal about them. ║
║ │ ║
║ │ // building the internet of ║
║ │ tomorrow ║
║ │ research software engineer ║
║ │ ║
║ │ applied AI research on the software ║
║ │ engineering side. trained models, ║
║ │ built the infrastructure to train ║
║ │ and measure them, and instrumented ║
║ │ existing systems to squeeze ║
║ │ research data out of them. ║
║ │ ║
║ │ what i shipped: ║
║ │ ║
║ │ LLM fine-tuning - supervised ║
║ │ fine-tuning of large language ║
║ │ models for structured, ║
║ │ domain-specific tasks. ║
║ │ curriculum-based data generation, ║
║ │ parameter-efficient training (LoRA) ║
║ │ on single GPUs and full fine-tunes ║
║ │ distributed across multi-GPU ║
║ │ clusters. wrote the eval harnesses ║
║ │ to actually verify the models ║
║ │ learned what they were supposed to. ║
║ │ ║
║ │ ML training infrastructure - ║
║ │ internal training-observability ║
║ │ tooling: thread-safe metric ║
║ │ tracking, distributed-aware ║
║ │ aggregation, anomaly detection, low ║
║ │ overhead by design. strict typing, ║
║ │ high coverage, CI. shipped. ║
║ │ systems instrumentation - ║
║ │ research-grade logging bolted into ║
║ │ existing high-performance systems ║
║ │ to quantify where the work was ║
║ │ actually happening, without ║
║ │ touching their behavior. ║
║ │ ║
║ │ infra - self-hosted internal ║
║ │ services and deployment automation. ║
║ │ ║
║ │ // research is just breaking things ║
║ │ on purpose and writing down what ║
║ │ fell out. ║
║ │ software engineer ║
║ │ ║
║ │ frontend and backend dev for ║
║ │ InCenter — an automated breach and ║
║ │ attack ║
║ │ simulation platform covering web, ║
║ │ network, and cloud security. ║
║ │ ║
║ │ deployed and configured InCenter ║
║ │ infra on AWS using terraform and ║
║ │ ansible. ║
║ │ ║
║ │ // the job where breaking things ║
║ │ was the actual job description. ║
║ │ network software engineer ║
║ │ ║
║ │ network software engineering on ║
║ │ proxy infrastructure at scale. ║
║ │ yes i thought about packets ║
║ │ constantly. yes that was fine. ║
║ │ founder & engineer ║
║ │ ║
║ │ solo founded and ran an AI ║
║ │ automation and marketing company. ║
║ │ just me. built everything, ║
║ │ sold everything, supported ║
║ │ everything. ║
║ │ ║
║ │ most of my clients were adult ║
║ │ content creators was on OnlyFans, ║
║ │ Fansly, Twitter, ║
║ │ Tumblr, Cam Sites, etc that whole ║
║ │ world. also had mainstream social ║
║ │ media growth clients and social ║
║ │ media marketing agencies work with ║
║ │ me to help their customers ║
║ │ (YouTube, Instagram, Spotify, ║
║ │ Discord, the usual). the adult ║
║ │ industry was the bulk of it. ║
║ │ ║
║ │ honestly the adult entertainment ║
║ │ industry pays better and the ║
║ │ technical problems are more ║
║ │ interesting. ║
║ │ ║
║ │ what i shipped: ║
║ │ ║
║ │ - stable diffusion pipelines + ║
║ │ custom AI image editing tooling, ║
║ │ you know, before "generative AI" ║
║ │ was a buzzword anyone used ║
║ │ - conversational AI agents handling ║
║ │ inbound DMs and fan interactions ║
║ │ ║
║ │ - full platform automation: ║
║ │ posting, scheduling, shares, ║
║ │ reposts, ║
║ │ likes, replies, follows running ║
║ │ simultaneously across Twitter, ║
║ │ Instagram, Tumblr and more, ║
║ │ hands-free ║
║ │ - automated transaction handling ║
║ │ fan pays, gets acknowledged, ║
║ │ money routes to the creator, zero ║
║ │ manual intervention ║
║ │ - audience targeting and growth ║
║ │ systems that turned one-time ║
║ │ followers into recurring paying ║
║ │ subscribers for my clients ║
║ │ - the whole stack was designed so ║
║ │ clients checked a dashboard ║
║ │ occasionally and otherwise forgot ║
║ │ it existed ║
║ │ ║
║ │ retention was the metric that ║
║ │ mattered. clients stayed because ║
║ │ their own subscriber bases kept ║
║ │ growing and paying. ║
║ │ recurring revenue on both ends. ║
║ │ ║
║ │ ran it all solo. customer ║
║ │ acquisition, support, dev, infra, ║
║ │ billing ║
║ │ ║
║ │ scaled it to the point someone made ║
║ │ an offer on the whole thing: ║
║ │ software, IP, client relationships, ║
║ │ resources, everything. ║
║ │ asset acquisition. ║
║ │ NDA covers buyer identity and ║
║ │ terms. ║
║ │ ║
║ │ // reverse engineered platform ║
║ │ APIs, built thin clients for major ║
║ │ social and messaging services, ran ║
║ │ adult content automation AI ║
║ │ infrastructure at scale. from my ║
║ │ bed. ║
║ │ "before it was cool." ║
║ │ embedded linux engineer ║
║ │ ║
║ │ built and compiled custom kernels. ║
║ │ ported linux to embedded devices. ║
║ │ wrote kernel-level and socket-level ║
║ │ code. developed and modified ║
║ │ firewall ║
║ │ and security software. custom linux ║
║ │ firmware with hand-applied patches. ║
║ │ ║
║ │ racked servers, NAS, switches. ║
║ │ configured DDNS, DHCP, name ║
║ │ servers. ║
║ │ debugged hardware issues. ran daily ║
║ │ internal and external pentests. ║
║ │ organized red and blue teams on the ║
║ │ company network. wrote security ║
║ │ reports and patching plans. ║
║ │ automated scanning of new database ║
║ │ leaks ║
║ │ for employee records. hired and ║
║ │ trained new employees. ║
║ │ ║
║ │ // if it compiles it ships. learned ║
║ │ why that's a bad philosophy here. ║
║ │ unix sysadmin / IT engineer ║
║ │ ║
║ │ domain user and group management. ║
║ │ rewrote linux and unix ║
║ │ documentation ║
║ │ that hadn't been touched in years. ║
║ │ ║
║ │ // the documentation was basically ║
║ │ archaeology ║
║ │ ║
║ │ resolved undocumented server issues ║
║ │ for clients and engineers. fixed ║
║ │ ICA remote client connectivity. ║
║ │ sorted out aruba router, VOIP, and ║
║ │ firewall configs. ║
║ │ rebooted the rdps; ║
║ │ ║
║ │ // resolved bugs that would have ║
║ │ led to over 6 million in ║
║ │ investments to disapear in under 5 ║
║ │ minutes ║
║ │ server engineer ║
║ │ ║
║ │ test servers, storage, switches and ║
║ │ sub-assemblies. ║
║ │ verify proper components are ║
║ │ received per order pick list prior ║
║ │ to build. ║
║ │ scan all components into the ║
║ │ manufacturing shop floor system to ║
║ │ ensure proper parts and inventory ║
║ │ tracking. ║
║ │ use online methods and work ║
║ │ instructions to build racks per ║
║ │ customer specification. ║
║ │ software engineer ║
║ │ ║
║ │ built internal tooling to automate ║
║ │ analytics and invoicing. ║
║ │ found an active RCE exploit a ║
║ │ former employee had left running. ║
║ │ killed it. ║
║ │ debugged a broken SMTP server. ║
║ │ built webservers for employee ║
║ │ training, ║
║ │ interview management, and ║
║ │ onboarding. ║
║ │ ║
║ │ // did incident response at a pizza ║
║ │ place. it counted. ║
║ │ tech support ║
║ │ ║
║ │ network monitoring, security ║
║ │ config, access permissions, ║
║ │ diagnosing ║
║ │ networking problems with diagnostic ║
║ │ tooling, routine maintenance. ║
║ │ ║
║ │ // early IT. everyone starts ║
║ │ somewhere. ║
║ │ IT support ║
║ │ ║
║ │ maintained schools networks, ║
║ │ internally and externally tested ║
║ │ and reported security flaws and ║
║ │ vulnerabilities district wide, ║
║ │ fixed teachers and student ║
║ │ machines, set up classroom networks ║
║ │ and devices. ║
║ │ ║
║ │ ║
║ │ // tldr: i got into 'ethical ║
║ │ hacking' by finding holes in ║
║ │ networks i was supposed to be using ║
║ │ to do homework on ;3 ║
║ │ // classic origin story ║
║ repair tech ║
║ ║
║ fixed arcade and vending machines. ║
║ soldering, wiring diagrams, ║
║ replacing ║
║ mechanical and electrical parts. ║
║ diagnosed malfunctions, ordered ║
║ parts, ║
║ handled billing. ║
║ ║
║ // this is where i learned that ║
║ hardware is just software you can ║
║ hit ║
╚════════════════════════════════════════╝
vxfemboy | meow <3 │
└────────────────────────────┘