What a profile reveals

One real example profile. The same page reveals more as a company moves from no account to a paid plan, then unlocks. Hidden fields are never sent to the browser, just removed.

  1. No account
  2. Company plan
  3. Unlocked

What anyone sees, signed in or not: the full achievements — the proof — plus skills, mission, domains, timezones, country and employment preferences. Completely anonymous: no name, no salary, no social links, and projects and in-public work show only their skills. Even the profile's public address is an opaque slug, never a name. A free account, talent or company, adds nothing to this.

p99 dropped from ~900ms to ~180ms; timeout rate fell below 0.1% — an 80% reduction sustained at 3x prior peak throughput.

See the full story

Backend Engineer · 8 yrs

Distributed systems & data platforms

Why is this profile anonymous?

Everyone here is anonymous by default — you see the proof, not the person. Their name, photo, and social links stay hidden until they share a private link, or a company unlocks the profile. Even then their e-mail is never exposed — companies reach them through messaging on the platform. It lets people share real work without their name attached, and stay in control of what they reveal, and to whom.

Proof of work

2026

Cut checkout API p99 latency by 80% under peak load

Under campaign traffic the checkout API degraded badly — p99 hit ~900ms, timeouts spiked, and customers were abandoning carts at the last step. I profiled the hot path instead of guessing, moved read-heavy lookups behind a write-through cache with event-driven invalidation, and added a covering index for the worst query. Each change shipped behind a flag with the old path one toggle away.

p99 dropped from ~900ms to ~180ms; timeout rate fell below 0.1% — an 80% reduction sustained at 3x prior peak throughput.

GoPostgreSQLRedis
2025

Designed an idempotent ledger — 0 double-charges in 12M txns

Retries sometimes double-posted transactions, so finance spent days reconciling and customers stopped trusting their statements. Live data had to migrate with zero downtime. I made every write idempotent behind a client-supplied key and rebuilt the balance model as an append-only ledger that is the single source of truth, with optimistic concurrency so safety did not cost throughput.

Double-charge incidents went to zero — 0 across ~12M transactions over two quarters.

GoPostgreSQLKafka
2024

Cut infra spend 52% by right-sizing a Kafka pipeline

The event pipeline was provisioned for a peak that never arrived — partitions and consumer groups sized off a forecast, burning budget every month with nobody owning the number. I measured real throughput and lag across a full cycle, repartitioned around the actual key distribution, scaled consumers on lag instead of a fixed count, and tuned retention to what teams actually replayed.

Monthly pipeline cost dropped ~52% with p99 consumer lag held under 2s — sustained through the next traffic peak.

KafkaGoKubernetes

I believe the best systems are boring on the surface and ruthless underneath — correct, observable, and quietly fast under load.

Off the clock

Locked
GoRedis
Locked
PostgreSQLTypeScriptGrafana

In public

System DesignCaching
Locked
Webpack
Locked

Practical bits

Timezones
Europe/AfricaAmericas EastAmericas West
Location
Poland
Languages
English, Spanish
Domains
Fintech, Banking
Employment
Full-time · Contract
Annual salary
Available with a company account
Social links
Available after unlocking

Reach out

Unlock with a company account to message directly and reveal:

  • Full name
  • Projects
  • In public
  • Social links
Unlock and contactSee what a profile reveals(opens in a new tab)

Find someone like this

Jump back to the board with this background already filtered in.

Backend EngineerGoPostgreSQLRedis
See the whole board

Which side of this are you on?

Get foundFind talent