/index · independent engineering practice

Software + security engineering for hard problems.

We build unusual systems, automate ugly workflows, connect things that were never designed to connect, and fix software that has become fragile, slow, or unsafe.

Colorado / available remotely across the U.S.

[ start a technical conversation ]

Recent work

all work →

2026-08 SAMPLE automation / software

Stop asking AI to click buttons it doesn't need to think about.

A screenshot-driven operations agent became a deterministic browser workflow, with model reasoning retained only where ambiguity required it.

61.7s → 4.8s median execution
92% fewer model tokens

2026-07 SAMPLE integration / software / automation

Three APIs. None agreed on reality.

A small integration layer reconciled incompatible models, incomplete APIs, and side effects that could not safely happen twice.

11 hr/week → <30 min manual reconciliation
0 silent sync failures after rollout

2026-06 SAMPLE automation / software

The demo worked. Production didn't.

An unstable AI prototype was rebuilt around bounded model behavior, explicit state, evaluations, and observable deterministic code.

p95 latency: 18.4s → 3.1s
inference spend: −71%

2026-05 SAMPLE security / software

Finding the bug was the first 20%.

An application-security review continued through patches, regression coverage, and verification instead of ending at a findings report.

3 critical / 7 high findings identified
all critical findings remediated before launch

2026-04 SAMPLE software / reliability

The safest rewrite was not rewriting it.

Instrumentation and behavioral tests made a fragile legacy service understandable enough to repair incrementally.

37 incidents/quarter → 2
MTTR: ~95 min → 11 min

Bring us this

problem index →

“It works, but nobody understands it.”
“We're doing this manually every day.”
“Our AI agent is slow and unreliable.”
“We found vulnerabilities. Now somebody has to fix them.”
“These systems need to cooperate.”
“The prototype needs to become a product.”

Capabilities

full index →

Software

applications / backends / APIs / integrations / legacy rescue / performance

Automation

workflows / browser control / AI systems / agents / evaluations / orchestration

Security

application security / architecture review / hardening / remediation / automation

/why

The person understanding the problem is the person working on it.

Pliska Labs is a founder-led independent practice. No account-management layer. No telephone game between a sales conversation and the implementation. Software, operations, and security remain one engineering conversation.

Technical notes

all notes →

2026-08-11
SAMPLE Browser agents should not reason about deterministic navigation
A model is useful when a step is ambiguous. Most interface navigation is not.
2026-07-24
SAMPLE The safest legacy rewrite is often no rewrite
Instrumentation changes a legacy-system decision from opinion into evidence.
2026-06-18
SAMPLE A useful security report has executable consequences
A severity label is not remediation guidance, and a finding is not finished when the PDF ships.

/contact

Send the messy version.

You do not need a complete specification. Explain what you are trying to build, fix, automate, integrate, or secure.

[email protected] →