A Carpathian valley from above; an AI scan turns the lake, trail and guesthouses into data
AI · reading the terrain
Summit
Trail to the summit
Lake · depth contours
Guesthouses
Spruce forest

I build with AI every day.

Not hype. Not a demo reel — a tool I reach for every day. It writes code beside me, and it ships as real features inside the products I build. Here is exactly how I use it, where it fits, and the rules I set for it.

Talk about your projectSee the projects
My AI loopAI does the typing — I own every decision
  1. Framea sharp prompt
  2. DraftAI proposes
  3. VerifyI review every line
  4. Shipto production
  5. Measurekeep or cut

Four ways AI earns its place

One tool, four very different jobs — from writing code beside me to features your users actually touch.

01 · In my terminal

As a pair engineer

Claude Code sits in my terminal as a second senior engineer — it reads the whole repo, follows my conventions, and never tires of the boring parts.

  • Scaffolds features, refactors, and framework migrations
  • Writes tests and reproduces bugs from a stack trace
  • Reviews diffs and explains unfamiliar code
  • Automates repetitive work — codemods, config, glue code
What I get
  • Days instead of weeks on repetitive work
  • More time on architecture and the hard calls
02 · In your product

AI inside the product

AI features users actually touch — search that understands meaning, assistants that answer from your data, call recordings turned into scored, searchable transcripts.

  • RAG & semantic search over your content
  • LLM features: assistants, scoring, parsing, extraction
  • Speech-to-text with speaker separation
  • Grounding checks so answers stay backed by sources
What you get
  • Features that feel smart, not gimmicky
  • AI that does a job — measurable, not a demo
03 · Content

Content & media, faster

AI removes the blank-page problem. Copy in three languages, social images, and video — all produced within the same visual system, not generic AI output.

  • Multilingual copy & SEO (EN / ES / UK)
  • OG images and brand visuals generated to spec
  • Short videos and showreels from scripts
  • This very page — drafted, then hand-finished
What I get
  • A blank page is never the bottleneck
  • On-brand output, produced in hours
04 · Routine

Automating the boring parts

Repetitive back-office work that used to eat hours — handled by small, reliable AI workflows that run themselves.

  • Ad-copy generation and bid rules (Google Ads)
  • Lead routing to Telegram / WhatsApp / email
  • Internal tooling and one-off data jobs
  • A team AI hub, so good solutions get reused instead of rebuilt
What you get
  • Hours back every week
  • Fewer manual, error-prone steps

AI I have actually shipped

Not slides. Not prototypes — real AI features running in production today. Explore any of them through the code or the live product.

CLCallLensSpeech + scoring

Sales calls turned into coaching insights — transcripts split by speaker, an LLM scorecard backed by quoted evidence, and every call searchable by meaning.

  • Deepgram speech-to-text with speaker separation
  • LLM scores each rep against a configurable scorecard
  • pgvector embeddings for meaning-based search
Symfony · Deepgram · OpenAI · pgvectorRepo + live
PPolylogHybrid RAG

A multi-agent chatbot SaaS with hybrid RAG (keyword + vector search, merged with RRF) and a verifier that blocks any answer not backed by sources.

  • bge-m3 embeddings on Elasticsearch
  • FastAPI backend, Next.js dashboard
  • A verifier checks every answer before it is sent
FastAPI · RAG · ElasticsearchLive
TQTripsQuickAssistant instead of search

An AI assistant that replaced site search — every answer grounded in our own depth surveys and public records, never the open internet.

  • e5 embeddings + kNN over measured facts
  • Result cards built server-side from Elasticsearch
  • Geocoding → driving route to the boat ramp
Next.js · Symfony · ElasticsearchLive
RCRybalka ClubAI assistant

A fishing community with an AI assistant that answers anglers’ questions from the community’s handbook.

  • e5-large embeddings for semantic search + RAG
  • Elasticsearch as the only datastore
  • Next.js + Symfony
Next.js · Embeddings · RAGLive

How I keep it honest

AI is powerful — and just as capable of being confidently wrong. These are the rules that keep it useful in production.

01

Human in the loop

AI writes the draft; I read every line before it ships. I own the decision, always.

02

Outcomes, not demos

If it doesn’t remove a real bottleneck or move a number, it doesn’t ship. Cool is not a reason.

03

Ground every answer

Answers stay tied to sources. A grounding check beats a confident hallucination every time.

04

Sometimes the win is removing AI

Sometimes the best AI solution is no AI at all. A static page or a simple rule can beat a model. Killing a feature is a valid result.

05

Own the data, pin model versions

Self-hosted where it matters, exact model versions pinned — no silent drift in production.

06

Fast to test, ruthless to remove

Prototype in a day, measure honestly, keep only what earns its place.

What I reach for

The models, retrieval, and plumbing behind the work above — chosen for production, not demos.

01Models & agents
ClaudeGPT (OpenAI)GeminiClaude Code
02Speech
Deepgramdiarization
03Retrieval
bge-m3e5-largepgvectorElasticsearch kNNRRF hybrid
04Orchestration
Symfony MessengerFastAPISymfony WorkflowRedis queues

Let’s put AI to work.

An AI feature, RAG system, or workflow worth automating — tell me the problem, and I’ll tell you whether AI is actually the right tool.

vitaly@kindrakevich.comDownload CV