Skip to content
SB

Shokhan Birlikov

@shokhan

ML Engineer interested in AI roles

Senior Data Scientist / ML / AI. Studied masters degree in Moscow State University. Live and work in USA.

Career

Experience: 7 years

Experience

Senior Data Scientist

CLARA Analytics

Provides AI-powered predictive analytics and decision-support tools for the insurance industry.

Mar 2025 — Present

Develops insurance prediction models and end-to-end ML workflows; built Claude Code skills that reduced some data-science workflows from hours to minutes.

• Develop and deploy ML models for insurance cost prediction, risk assessment, and classification, including feature engineering, tuning, and threshold optimization. • Built internal Claude Code agent skills for the data-science team - automated EDA, a Confluence report publisher, and a conversational natural-language interface to the Athena data warehouse - cutting workflows that took hours down to minutes. • Use AI coding agents (Claude Code) in day-to-day engineering for code development, merge-request reviews, and report writing.

Data Scientist, ML Development

Behavox

Provides AI-powered compliance and communications-monitoring software for businesses.

Jan 2023 — Apr 2025

Built multilingual text-classification and compliance models, including language identification and signature tagging; also supported ASR training pipelines.

• Developed text-classification models and compliance applications for detecting suspicious communications and bad actors across multiple languages and communication channels. • Trained and fine-tuned a range of models - linear classifiers, BERT transformers, and Large Language Models - with extensive result analysis (threshold selection, dataset preparation, parameter tuning). • Supported Automatic Speech Recognition (ASR) projects through dataset preparation and running training pipelines. • Developed Language Identifier and Signature Tagger models.

Senior Machine Learning Engineer

Galamat-Tech

Mar 2022 — Jan 2023

Built conversational AI for 10+ local businesses, an NLU service for a chatbot platform, and language-processing tools.

• Developed conversational AI and goal-oriented chatbots deployed for 10+ local businesses; worked closely with business to convert requests into technical tasks, and owned development and maintenance. • Built the Natural Language Understanding service for a Chatbot Constructor platform - semantic question-answering and retrieval-based context grounding for more accurate chats; ran experiments with open-source language models.

Machine Learning Engineer

BTS-Digital

Oct 2019 — Mar 2022

Built bilingual virtual-assistant and chatbot systems, face anti-spoofing for a national platform used by 8M people, and an OCR document scanner.

• Built a bilingual (Kazakh-Russian) virtual assistant and an intent-based dialogue system. • Developed goal-oriented chatbots and automated call-center assistants helping local businesses automate support and sales. • Researched and implemented Face Anti-Spoofing algorithms for the "Digital ID" face-biometrics system used by 8M users of a national gov-tech platform. • Built a custom OCR-based document scanner tool from scratch.

Machine Learning Engineer

AI-Labs

May 2019 — Sep 2019

Prototyped an audio-to-lip-animation project and contributed to a National Bank virtual assistant serving 100K online users.

• Built "Speech2Lips" from scratch - an interdisciplinary project generating lip animation from audio speech; prototyped and validated application viability. • Contributed to a Virtual Assistant chatbot for the National Bank serving 100K online users.

Education

M.S., Applied Mathematics and Computer Science

Moscow State University

2017

M.S., Machine Learning

MADE Data Science Academy

2021

B.S., Applied Mathematics

Aktobe State University

2013

Selected projects

Zadrz

Led development of an AI-powered messaging platform with personalized conversational agents ("AI clones"), owning system architecture, LLM integrations, and cloud infrastructure for scalable, real-time, knowledge-aware interactions.

KazLLM

Contributed data preparation for a local generative foundational LLM for the Kazakh language: built de-duplication, filtering, and cleaning scripts for SFT datasets, and prepared a CommonCrawl dataset following the FineWeb recipe.

AI-Legal / Saraptama-AI

Built legal-tech tools: a platform retrieving and recommending relevant legal issues using summarization, semantic similarity, and information retrieval; and an automated contract analyzer flagging problematic clauses with suggestions.

Booketition

Built an automated question-generation pipeline for a reading-contest platform using models for summarization and question generation, with text processing and feature extraction.

Achievements

Top-1% in mlcourse.ai

OpenDataScience · 2019

Lomonosov Math Olympiad - 1st place

2015

Republic Math Olympiad (undergraduate, Kazakh National University) - 2nd place

2010

Republic Math Olympiad (undergraduate, Kazakh National University) - 3rd place

2011

Skills

PythonPyTorchTensorFlowHugging FaceScikit-LearnLLMsRAGprompt engineeringagent orchestrationLangChainvector databasesClaude CodeMCPtool/skill developmentNLPConversational AINERintent classificationRasaASR tooling (Kaldi K2, Lhotse) - exposure