PETR NOVIKOV
AI STARTUP MENTOR
Helping businesses leverage the power of Artificial Intelligence.
WHY ME
Experience in a variety of roles ranging from Frontend Developer to Tech Lead.
Diverse experience at leading IT companies, including Apple Inc.
Proven record of building AI projects, from idea pitch to user satisfaction.
01.
02.
03.
Professional Math teacher in the past, which includes teaching in high-profile families and at a US university.
Efficient dealing with uncertainties, limited budgets, and working under pressure.
04.
05.
WHY YOUR AI PROJECT
NEEDS A STARTUP MENTOR

Starting a new tech project is always a major challenge. It's hard to come up with a feasible business idea, convince the investors your idea is worthy, build the actual product and convince the customer to buy your product, juggling the risks and uncertainties all the way in highly competitive environment. Putting Artificial Intelligence in the core of your business normally makes your project more attractive and visible, but also brings in additional challenges and uncertainties, those conditioned by non-deterministic nature of AI itself, the fact that at times building an AI product feels more like scientific research than software engineering, and the difficulty of distilling spectacular, but not so readily usable "AI magic" into tangible customer value.


When it comes to hiring strategy, it's tempting to say you need the top talent for everything, but those are normally very busy and often not readily available. The other extreme, i.e. to invite whoever is available and hope that their ambitions, self-motivation, and hard work alone will yield the desired results, is often infeasible in reality. The best is balancing the two: to hire, or partner with, one or two top-notch professionals so they can guide and teach their less experienced colleagues, ensuring great overall quality of work at the same time.


The benefits of having an ultra-experienced member in your team are obvious: this raises the quality of decision-making, technical and business alike, enables faster adoption of best practices, allows to achieve results in less time as more senior professionals use to see shorter paths and know what works and what doesn't, and, to some extent, gives the team peace of mind, as they feel they won't be left alone if something goes wrong, and they always have somebody to learn from. All in all, in the context of fast-paced business environment, it is better to rely on somebody's existing experience rather than expecting everybody to gain their own experience through hardships and failures.


The important thing is that such experts not only must be the ones who can do the things on their own, but also be able to transfer their precious knowledge to the rest of the team. This is not normally efficiently achieved by mere lecturing or giving out ready solutions, instead, it's best if the expert will teach and guide the team inspiring independent approach and responsible attitude, much like best school and college teachers do, except it's a budget- and time-constrained set-up. Much like business management, teaching is an art of its own, and those who master teaching alongside their professional skills make great mentors. A carefully chosen mentor can save your project a lot of time and money, improve working comfort for the founders and the team, and increase the overall viability and competitiveness of the project as a whole.


As an experienced AI developer with substantial research, teaching and mentorship background, I am open to opportunities of becoming a mentor for a project that I'll find interesting. My experience so far includes the applications of AI in Software Security, Supply Chain Management, Predictive Maintenance, Manufacturing Quality, Operations Research, and Computer Vision, and I am open to collaboration in virtually any area. Please learn How I Can Help, explore the details of my past experience in Case Study section, alongside Feedback And Recommendations from my colleagues, and, in case you are interested, feel free to Contact Me.



  • Extracting the business value from AI
  • Identifying the right business problem to be solved by the project
  • Finding out what exactly the customer will be ready to pay for
  • Crafting proper success metrics
  • Identifying the right goals and objectives
  • Building the path to the desired results that takes minimum time and efforts
  • Robust planning: knowing what to do if something goes wrong
  • Tackling the risks and uncertainties
HOW I CAN HELP

1. BUSINESS MODEL
  • Showcasing the product
  • Identifying the real customer needs that our product can solve
  • Efficient pre-sales
4. CUSTOMER RELATIONS
2. ROAD MAP
3. ARCHITECTURE & STACK
  • Overall system architecture, coupled with plans and actual capabilities
  • Choosing the right programming languages and libraries
  • Choosing proper AI and ML models
  • Making the maximum of available computational resources
  • Building MVP fast
  • Highlighting the value of investing in this project
  • Convincing the investors we are a great team
5. INVESTOR RELATIONS
6. TEAM NURTURING
  • Identifying where to find the right people
  • Identifying the value that our project bring to the the team members, besides paychecks
  • Organizing learning without interrupting the working process
CASE STUDIES
Case Study: Apple Inc.
1. ADAPTIVE QC FOR ELECTRONICS MANUFACTURING
AI-powered quality control system for factory assembly line process optimization

Being the world's leading consumer electronics manufacturers, Apple would not compromise on the quality of their products, but at the same time seeks to cut manufacturing costs, like every other business.

The outstanding manufacturing quality is normally achieved through the use of mathematically-intensive process improvement methodologies, such as Six Sigma and Lean Manufacturing, whose combinations, although providing the desired output quality, may still have some room for improvement. After demonstrating potential cost reduction through POC, a necessity for somebody who will take on the "productionalization" of the POC arised, resulting in Apple hiring me as a contractor (through Grid Dynamics).

In this project we used Data Analysis, Machine Learning, Genetic Algorithms, and Six Sigma Methodology to build an internal tool that allowed hardware engineers to automate the adjustment of quality control criteria at multiple points on the assembly line, reducing the number of hardware components discarded while keeping the invariably high quality of the final assembled products, thus providing notable cost savings.

The system was later introduced in many of Apple products' manufacturing assembly lines, ranging from peripherical devices to flagman products.
Case Study: Astra Linux
2. LLM FOR LINUX OPERATING SYSTEM SECURITY AND COMPLIANCE
Building an LLM-powered software security assistant for Astra Linux operating system development pipeline

With software vulnerability mitigation being top priority for a high security operating system such as Astra Linux, millions of lines of code for inspection and tight shipping schedule, a tool that enables the software security experts to automate their work and keep up with the ongoing code changes and updates would be handy. As an experienced ML expert, I was invited to assume a leading role in the development of such automation tool.

In this project we developed the automation tool for code inspection facilitation and integrated it in Astra Linux's development pipeline, which allowed for more than 50% in saved time and made the regulatory compliance easier and more efficient. The system has Large Language Models at its AI core, which, due to the obvious security concerns, was adapted to run in on-premise settings.

The tool was later accepted as standard not only in the Astra Linux OS development, but also across many other Astra's products, especially those that are security-critical and compliance-heavy.
Case Study: Lean Technologies LLC
3. LEAN MANUFACTURING STARTUP FOR HOUSEHOLD CHEMICALS PRODUCTION
Helping jumpstarting a new business in Lean Manufacturing services

As two of my friends embracing the opportunity for collaboration with a large household chemicals manufacturer and needing somebody who will provide his relevant experience reached out to me, I was happy to help them get the things started.

Under or largely under my guidance, we identified the yet-to-be customer needs, created multiple business proposals, largely based on my prior experience with Apple Inc. and Applied Logistics, and ultimately the newly established company reached a partnership deal with Nefis Group, one of Russia's largest household chemicals manufacturers. The pilot project was deemed successful and collaboration was established.
Case Study: Applied Logistics
4. SUPPLY CHAIN OPTIMIZATION IN MACHINE MANUFACTURING
Building supply chain for machine industry production needs

As Applied Logistics embraced the opportunity to design a multi-layer geographically dispersed supply chain for a major self-propelled machine manufacturer, they found themselves in need of expertise to put together advanced mathematical methods, customer financial feasibility and the reality; thus they hired me.

Using Operations Research, Reliability Theory and Linear Optimization, we built a model that calculated optimal geographical position and service hubs' stock levels, given the machines' maximum allowed downtime.

The solution enabled the customer to transfer to a new service model and to fulfill new regulatory requirements and contract obligations. The solution was shipped to the customer as a plugin to Applied Logistics’ flagman software product.
Case Study: Abiroy | Astra Linux
3. SMART DATA DEDUPLICATION
Читать далее
Case Study: RoboCV
5. COMPUTER VISION FOR SELF-DRIVING VEHICLE STARTUP
Helping early stage self-driving vehicle startup pass a crucial financial stage.

Being hot topic at the time, startups aiming for self-driving cars looked promising and attractive for investment. RoboCV's promise was not the vehicle proper, but an autonomous navigation system that could be installed on any vehicle which used computer vision and advanced mathematical methods for navigation and decision-making.

As I joined RoboCV, I found myself in the turmoil of an early stage business. Being a team of just five, we had to take on both engineering and non-engineering duties, including marketing, public and media relations, investment-related paperwork and desperate search for new funding.

Despite the early stage difficulties, we successfully made a few sale and investor pitches, pivoting for warehouse vehicles, resulting in further funding. The startup later grew into one of Russia's leading companies in warehouse vehicle autopilot systems.

FEEDBACK AND RECOMMENDATIONS

PROFESSIONAL BIOGRAPHY
  • 2023–present: Lead Programmer at Astra Linux
  • 2023–2024: Co-founder at Lean Technologies Laboratory LLC
  • 2020–2022: Machine Learning Engineer at Apple Inc. (via Grid Dynamics)
  • 2019–2020: Lead Research Engineer at Diginavis
  • 2018–2019: Lead Programmer at Abiroy
  • 2015–2018: Lecturer at Kazan Federal University, IT department
  • 2013–2014: Business Analyst at Applied Logistics
  • 2012-2013: Research Programmer at RoboCV
  • 2009–2012: Self-employed tutor of Mathematics, Programming and Physics.
  • 2010: PhD in Mathematical Statistics from Moscow State University
  • 2008: Graduate student/Teaching Assistant at the University of Toledo (Ohio, USA).
  • 2005: Graduated from Kazan Federal University (Kazan, Russia), Faculty of Mechanics and Mathematics.

See Full CV >> (LinkedIn)
Tilda Publishing
CONTACT A TECH-STAR STARTUP MENTOR
TO LEVERAGE THE POWER OF AI
If I see the potential in your idea, I will be delighted to help you fully unleash it, helping you to minimize the risks and increasing the chance of success. So far my experience includes the following areas:

  • AI in Manufacturing
  • Supply Chain Management
  • Predictive Maintenance
  • Quality Control
  • Software Security
  • Large Language Models
  • Operations Research
  • Optimization
  • Computer Vision
  • Business Analytics

Please fill the form below so that I could reach back to you and discuss the details of potential collaboration.
Please notice that I do not expect remuneration during the acquaintance/discussion phase.






Contact me
Please fill in your data so I can reach back to you
What are your exact topics of interest?

© aiforstartups.ru, 2025
© freepik.com, illustrations