00 / index
built to stay understandable.
a quiet corner of the internet for work in progress, technical notes, mistakes worth remembering, and the occasional odd idea.
- learning
- rust, slowly
- building
- esshi.dev
- running
- too many small servers
type help or click a directory · tab completes, ↑ recalls
01 / projects
systems, strategy, and things being built.
led technical discovery for a service cloud case-duplication problem spanning salesforce and several external service platforms. mapped current and target flows, defined system-of-record ownership, case-id propagation, idempotent behavior, exception paths, and a customer-owned implementation plan later sharpened through production validation.
designed a regional seller-enablement motion to identify post-sale risk earlier, activate the right customer success resources, and improve adoption outcomes. developed targeted leadership outreach, practical customer scenarios, and the finished enablement deck; delivery sessions are now scheduled.
a rust-powered platform for configurable business software: flexible schemas, structured records, permissions, automation, versioned configuration, and controlled releases. the current work is defining the core object model, module contract, and cargo workspace before implementation begins.
built and hardened a self-hosted linux environment for containerized services and multiplayer infrastructure. the system covers docker workloads, systemd services, private administration through tailscale, firewall policy, ssh hardening, service-user separation, vulnerability scanning, logging, health checks, and recovery procedures.
designed an application-aware off-site backup workflow that coordinates clean shutdown, encrypted snapshots, remote storage, restart, health checks, and representative restores. restored files are compared against production with sha-256; identity drift and failed validation stop the workflow instead of producing a misleading green checkmark.
an llm-powered operations agent that translates natural language into controlled actions against a linux environment. execution stays behind deterministic tools, allowlists, confirmation flows, narrow sudo boundaries, and separate conversational and privileged components—never unrestricted shell access disguised as ai automation.
this website: a public home for projects, notes, and ideas, built to understand the complete path from source code to a server on the internet. static by choice, versioned on github, built on the host, and served through a small debian and nginx stack with fewer moving parts to maintain.
02 / notes
field notes from figuring things out.
nothing published yet.
the current margins contain rust, self-hosting, infrastructure ownership, small technical discoveries, and things i broke so i would understand them.
03 / about
between business problems and technical systems.
i’m ian. i work best where business problems, people, and technical systems meet. i like understanding what someone is trying to accomplish, finding where the system creates friction, and turning that into a clearer technical direction.
- role
- success engineer, cloud success
- based
- mexico
- languages
- english / spanish
- learning
- portuguese / rust / infrastructure
professional work
enterprise delivery, systems, and customer operations.
aug 2026 — now
success engineer / cloud success
salesforce.com ↗
i work with executives, it leaders, administrators, and developers across premier and signature customers. the job moves from platform assessment and success planning into technical discovery across service cloud, agentforce, cti, third-party ai agents, automation, api and middleware flows, and the broader salesforce platform. i turn production behavior, business constraints, and stakeholder priorities into a sequenced technical direction.
dec 2025 — aug 2026
onboarding specialist / cloud success
i owned a structured kick-off, discovery, and recommendations lifecycle for enterprise service cloud customers across north america and latam. i managed up to 25 concurrent engagements while maintaining a 4.8+ csat, ran reverse-demo health checks across case intake channels, and authored a repeatable discovery framework based on account forensics, evidence quality, and decision-driven questions.
aug 2025 — dec 2025
customer support / crm case operations
i owned customer cases end to end in salesforce, maintaining the documentation and audit trail across each lifecycle. i served as a primary contact across more than 450 cases per week while sustaining quality scores near 99%, coordinating cross-functional resolution through slack and confluence, and documenting reusable answers for the wider team.
sep 2024 — jul 2025
enterprise account operations / business intelligence
i managed hundreds of enterprise accounts representing millions in billing while owning the reporting layer for a 90-agent operation. i built self-refreshing kpi dashboards for daily leadership decisions, then scoped, configured, and demonstrated a power bi migration for broader distribution and deeper analysis. data-driven adjustments also moved my own performance from 85% to 105% in under three months.
apr 2024 — oct 2024
international service operations
i was a top performer across international service centers in a newly launched vertical, consistently exceeding quality and efficiency targets while feeding frontline observations back into team alignment during the ramp-up.
feb 2023 — apr 2024
customer support / de-escalation
i handled more than 60 time-sensitive support issues each day, reducing higher-tier intervention through clear expectations and practical resolution. i also mentored teammates on de-escalation technique and helped raise team productivity.
data & presentation
useful should never mean ugly.
i’m good with numbers and, against my better judgment, genuinely love excel. power query is my favorite feature: i like taking messy sources, building reliable transformations, and turning a pile of tables into a data model people can actually use.
i have a thing for translating data models and queries into clear, useful dashboards. the same rule applies to presentations and everything else i ship: it should work, it should make sense, and it should not be an eyesore. i dare you to find an ugly presentation with my name on it.
engineering practice
learning the whole system.
outside work, i’m designing rusty schema, building this site, and operating my own small servers. together they cover data modeling, modular architecture, rust, web development, linux, networking, security, deployment, backups, and technical documentation.
the common thread is ownership of the complete path: from an ambiguous problem, through a model and implementation, to the infrastructure that keeps it running. i would rather understand a system imperfectly than treat it as a magic box.
what matters
capability should be shared.
i care about who owns the systems we depend on, who performs the work, and who ultimately benefits from them. i’m optimistic about technology, but skeptical of artificial scarcity, permanent lock-in, and progress measured only through private accumulation.
i’m interested in open knowledge, accountable infrastructure, broadly accessible tools, and automation that returns time and security to people. progress should mean greater capability and abundance without concentrated power.
credentials
- salesforce certified advanced administrator
- salesforce certified platform app builder
- salesforce certified administrator
04 / contact
say hello.
for work, systems, strange ideas, or a sufficiently interesting problem.