Lev Sklyarenko

Business & systems analyst · Capital markets · FinTech

I take financial products from specification to production – and stay responsible for them afterwards. Four derivatives live at Sber CIB with a fifth in progress; before that, treasury automation at VTB and budgeting systems for LUKOIL at Kept.

The through-line is the same in each: understand the economics, design the system behaviour, ship it, then own the integrations and reconciliations that follow. Python and SQL are working tools here, not line items – and the two valuation models further down are mine, open in your browser, formulas and all.

LinkedIn Telegram Instagram Gmail Phone

Where I have worked

Three institutions, one thread.

A Big-4 advisory practice, a bank treasury, and an investment bank – each one closer to the instrument than the last. 2.5+ years, the first eighteen months of them alongside a full-time degree.

May 2025 → now

New Products Cluster. Four derivatives taken live in Murex, a fifth in progress.
Sep 2024 → May 2025

Treasury. Excel and Power Query workflows rewritten in Python for five internal clients.
Feb → Sep 2024

Advisory & Digital CFO, ex-KPMG. Budgeting transformation for LUKOIL in SAP HANA.
Independent

Oct 2023 → May 2025 · alongside

Freelance back-end, taken on next to everything else. Five delivered Python services and Telegram bots.

The hollow stop is freelance work: it ran alongside the first two jobs rather than between them.

Eighty percent of success is showing up.

Woody Allen

Track record

Experience

What I actually do, job by job.

May 2025 → present · Moscow

Business / systems analyst, product team

Embedded in a cross-functional team serving the CIB trading desk. I own the end-to-end delivery lifecycle for new capital markets instruments inside Murex.

Delivered

Four products live in Murex production – an exotic option embedded in a structured note, plus deliverable and cash-settled OTC client options on equity and bonds under separate accounting models. Business requirements through UAT to handover, on each one.
Took a key cross-system reconciliation from 100–200 daily breaks to zero by reworking the data-processing logic behind it. A second, payment-side reconciliation is down from ~5,600 rows to ~2,300 and still falling.
Automated a team workflow on my own initiative – database extractions that took 15 minutes to over an hour by hand now take seconds. It is being scaled to cluster level.
Unified the product-flag enrichment methodology across all 82 team products and documented every attribute in Confluence, so there is one source of truth instead of tribal knowledge.

Running now

Two streams in parallel as the analyst: a new FRA on the spread between two rates rather than a single rate, and the revival of three exotic options – binary, Asian and barrier OTC equity – to product-team standards.
Leading a daily cross-system reconciliation initiative for structured notes, and designing its automation as an AI-agent application.
FORTS, SPFI and other OTC instruments; advising traders and adjacent systems on Murex; production support.
The team product catalogue and business-process documentation in BPMN, delivered under Scrum and Agile 2.0.

Sep 2024 → May 2025 · Moscow

Financial model automation specialist, Treasury

Python automation for treasury workflows – requirements gathering to deployment, run independently, concurrent with my BSc.

Migrated Excel macro and Power Query workflows to Python and Lua for five internal clients across treasury departments – on average 60–80% less manual processing time per task.
Supported the OS migration from Windows to Astra Linux by replacing VBA macros with cross-platform Python.
Worked with the VTB Capital and VTB CIB divisions, which is where treasury stopped being an abstraction to me.
Ran the stakeholder meetings myself, measured the effect of each automation, and wrote the onboarding guides people actually used.

Feb 2024 → Sep 2024 · Moscow

Consultant, Advisory & Digital CFO

Budgeting transformation and unification for LUKOIL – automated reporting in SAP HANA replacing a manual Excel process. Also concurrent with the degree.

Built the financial models and budgeting methodology to the client’s RAS requirements and handed them over after UAT.
Modelled and documented the new reporting process architecture in BPMN.
Ran budget analysis and plan-versus-fact variance analysis, identifying discrepancies across the reporting structure.
Produced the stakeholder-facing roadmap presentations for the initiative milestones.

What I work with

Skills

Every keyword from my CV, and where I actually used it.

Filter by category, then click any tag. The panel underneath says what I did with it – nothing here is listed for decoration.

Pick a taghow this works

Each one comes from my CV. Click it and this panel says where it was used.

My models

Models

Two valuations I built, sheet by sheet.

Both were built in Excel for my degree, and between them they carry 7,705 formulas across 27 sheets. What follows is how each one is put together – the structure and the reasoning, with the sheets themselves alongside.

LUKOIL – valuing an oil major from its cash flows

A five-year cash-flow forecast to 2029, cost of capital built from the ground up, two independent terminal values, and a peer cross-check on EV/EBITDA.

₽7,680

implied share price the model lands on

17.4%

weighted average cost of capital

3,964

formulas across its fourteen sheets

LUKOIL Financial Statements sheet: inflation and GDP forecasts with links to the Central Bank, revenue and cost indexation, and the P&L down to gross profit — Financial Statements **Assumptions with a source attached.** Russian inflation and GDP growth are pulled from the Central Bank’s own forecast – the links sit in the row labels – and revenue and costs are indexed off them at 3% a year. Gross margin drifts from 48,5% down to 42,6% as a result, rather than being typed in.

LUKOIL DCF sheet with the formula bar showing =XNPV(D20;D18:H18;D7:H7), EBITDA, FCFF and discounted cash flow for 2025 to 2029, a 1.44x exit multiple, two terminal values and a 17.4% WACC — DCF **Discounted against real dates.** The formula bar is the point: `=XNPV(D20;D18:H18;D7:H7)` discounts to actual year-ends rather than whole periods. Both terminal values sit side by side – 1 642 344 on a 1,44× exit, 1 229 003 on a Gordon perpetuity – and neither is quietly preferred.

Two sensitivity tables for LUKOIL: implied share price and implied enterprise value, each across WACC from 15.4 to 19.4 percent and exit multiples from 1.0x to 1.8x — DCF, sensitivity **The answer as a range.** Share price against WACC and exit multiple runs from ₽7 503 to ₽7 874 – a spread of about 5% across the whole grid. Worth knowing before you argue with the base case: on this model the discount rate moves the answer far less than you would expect.

LUKOIL implied value: enterprise value of 3,480,513 from the DCF and 4,663,961 from comparables, averaged to 4,072,237, less debt plus cash to give equity value of 5,321,479 over 692.9 million shares, or 7,680 roubles per share — DCF, rows 48–57 **Down to one number, and you can see every step to it.** The DCF gives an enterprise value of 3 480 513; the peer set gives 4 663 961; the model averages them rather than picking a favourite. Net off debt, add the cash balance, divide by 692,9mn shares – **₽7 680 a share**. The toggles on the right switch the terminal-value method and the cost-of-equity approach.

Click any sheet to enlarge it. These are screenshots of the workbook as it stands, not redrawn for the web – the formatting, the row labels and the formula bar are all as I built them.

01

Rebuild the history before forecasting anything

Three statements from the IFRS filings, tied together, so every forecast line has a real base to be checked against.

02

Keep the cash-flow build on its own sheet

The Calcs sheet walks EBITDA down to free cash flow. The valuation sheet only ever references it – =Calcs!F193 – so the calculation lives in exactly one place.

03

Build the discount rate, don't assume it

Cost of debt is the borrowing rate taxed at the effective rate off the accounts; cost of equity adds a country risk premium to a risk-free rate. The capital structure weighting them comes off the projections rather than being typed in.

04

Discount on real dates

The model uses XNPV against actual year-ends rather than integer periods. At a 17% discount rate the day count is worth real money, and 2028 is a leap year.

05

Take the terminal value two ways, and don't pick a winner

One leg exits at an EV/EBITDA multiple, the other runs a Gordon perpetuity. The base case averages them, because neither method is right on its own.

06

Finish with a range, not a number

A six-name Russian oil & gas peer set gives a second enterprise value, and the two-way sensitivity tables on the sheet show how far the answer travels with the discount rate.

What this model is and isn't. It was built for my degree on public filings and market data. The share price is what the model produces under the assumptions visible on the sheet – not a recommendation, and the cost of capital behind it is a point estimate rather than a range.

Magnit – building a retailer's revenue from the basket up

The more complete of the two. Revenue comes from average ticket and customer traffic rather than a single growth rate, then runs through a full three-statement projection to 2030 and into the DCF.

₽4,471

implied share price, against ₽3,234 in the market

14.2%

weighted average cost of capital

3,741

formulas across its thirteen sheets

Magnit Projections sheet: revenue split into retail, wholesale and other from 2019 actual through 2027 forecast, with cost of sales below — Projections, rows 12–17 **Revenue split three ways, six years back and six forward.** Retail carries 97–99% of it, wholesale has been shrinking since 2022, and “other” only appears in 2023. Keeping the three lines separate is what makes the forecast arguable in a useful way.

Magnit Assumptions sheet: revenue growth, average ticket growth, traffic growth, COGS and SG&A as a percentage of revenue, other income and finance costs, for 2020 to 2030, with CAGR and change in basis points — Assumptions **Revenue from the basket up.** Growth is average ticket times customer traffic, not one blended rate – ticket at 7,9–8,6% a year against traffic at 0,7–0,9%, which is a specific, checkable claim about a grocer. Every cost line is held as a percentage of revenue, and the right-hand columns carry the CAGR and the change in basis points so drift is visible.

Magnit DCF sheet: EBITDA and working capital, net income bridged to free cash flow to firm through interest, tax shield, minority interest, depreciation, working capital and capex, a 1.5 percent long-term growth rate, Gordon terminal value and a 14.2 percent WACC — DCF Valuation **Net income walked all the way to free cash flow.** Interest added back, the tax shield on it taken off, minority interest, D&A, the working-capital move and capex – each on its own row, so the bridge can be audited line by line. Terminal value on a 1,5% perpetuity, discounted at **14,2%**.

Magnit Projections sheet: the full profit and loss statement from revenue to profit for the year, 2019 to 2030, including operating profit, interest, tax, profit attributable to shareholders and non-controlling interests, and EBITDA — Projections, full P&L **Twelve years on one sheet, and almost all of it formulas.** Revenue to profit for the year, with six years of history in grey to hold the forecast honest and the split between parent shareholders and non-controlling interests carried properly. Revenue runs 1 368 705 to 5 137 098; EBITDA 147 310 to 397 771. This one sheet holds **1 091 of the workbook’s formulas**.

Click any sheet to enlarge it. The blue columns are forecast, the grey ones are reported history – the colour convention runs through the whole workbook.

01

Start with what management actually controls

Revenue is average ticket growth times traffic growth, split across retail, wholesale and other. That makes the forecast arguable in the right way – you can disagree with the traffic assumption specifically.

02

Anchor every ratio to six years of history

The Ratios sheet covers liquidity, activity, profitability and coverage for 2019–2024. If a forecast margin drifts outside where the business has ever operated, that sheet catches it.

03

Project all three statements, tied

P&L, balance sheet and cash flow to 2030, linked so the balance sheet has to balance. Most of this model's formulas live on that one sheet.

04

Be explicit about the capital structure

The debt weight comes out at 81%, which looks extreme until you see why: this is a grocer that leases its stores, so IFRS 16 lease liabilities are 63% of total debt, and the equity leg is book value rather than market cap. Both choices are on the sheet, not buried.

Credit where it is due. This model was co-authored – the workbook is titled “by 2 Levs” – and the revenue build, projections and DCF were split between us. Worth flagging too: the share count carried in the projection is 67.85mn against 97.55mn issued historically, so the per-share figure depends on that basis.

It is better to be vaguely right than exactly wrong.

Carveth Read – Logic, 1898 · the line usually misattributed to Keynes

Education & credentials

Education

Still studying, deliberately.

MSc, Financial Markets

Sep 2025 → Jun 2027

HSE University – Faculty of Economics, FMFI.

Current project: M&A deal effectiveness in pharma and biotech, 2015–2025.
Current project: cost-management and risk-assessment tooling for credit institutions, industrial enterprises and alternative finance providers.

Financial University under the Government of the Russian Federation

BSc, International Finance

Sep 2021 → Jun 2025

Financial University under the Government of the Russian Federation – Faculty of International Economic Relations.

Graduation startup project: a fair-value calculator for derivative instruments with AI integration.
VAK-indexed research paper: risk-management adaptation for modern markets – a Sberbank case study, 2022–2024.

Japanese, rare languages course

Sep 2021 → Jun 2025

Moscow State Institute of International Relations (MGIMO) – four years, in parallel with the finance degree.

Graduated 86/100, at roughly A2–B1 proficiency.

Alongside

SF Education – CFO and Investment Advisor programme, in progress.
Harvard CS50 – Introduction to Computer Science, 2023.
Alfa-Bank – Product Management Technologies, product owner track, 2023; finalist in the Alfa-Bank case championship.
Changellenge Cup Moscow 2023 – High Quality Award, top 15%.

Away from the desk

About

The part that isn’t on the CV.

I was meant to be a twin. Early on in development I apparently decided against the idea – and this is what came of it.

Cheerful, and I would like to think kind. I love photographing nature, places worth stopping for, and being in loud company. I travel with my girlfriend: so far that is Kota Kinabalu in Malaysia and Zanzibar, which is a short list on purpose – it has only just started.

There is a cat and a dog at home. They do not fly with us yet. They will.

Something is nearly always playing: soul, britpop, rock. There is a shelf of it on the Music page.

Football

Barcelona

Tennis

Carlos
Alcaraz

Painting

Ivan Aivazovsky Gustav Klimt Wassily Kandinsky Henri Rousseau Nadezhda Udaltsova

Music

Soul Britpop Rock

Four I keep coming back to

Ivan Aivazovsky, The Road from Mleta to Gudauri: a carriage on a cliff road above a fog-filled Caucasus gorge, snow peaks beyond — **Ivan Aivazovsky**
The Road from Mleta to Gudauri, 1868

Wassily Kandinsky, Composition VIII: circles, triangles and straight lines in primary colours on a pale ground — **Wassily Kandinsky**
Composition VIII, 1923

Henri Rousseau, A View of the Bridge at Sevres: a stone bridge over a river with a balloon, an airship and a biplane in the sky — **Henri Rousseau**
A View of the Bridge at Sèvres

Nadezhda Udaltsova, Nov: a cubist still life in ochre, grey and blue with fragments of Cyrillic lettering — **Nadezhda Udaltsova**
“Новь”

Photography

These are my photographs of nature and of the places I have been. I try to catch the moments that carry some feeling in them. Travel, for me, is a way to open up new horizons and to meet the culture and the people of another country.

A single tall cloud lit gold by the setting sun above a calm sea, heavy dark cloud to the right — **Pl. 02**One cloud getting all of the light.

A minimal frame: flat blue sea below an even blue sky, the horizon a single line — **Pl. 16**Two blues and a horizon.

The sun touching the horizon in a golden sky, rows of paddleboarders in silhouette on the water — **Pl. 04**The sun going down behind the paddleboards.

Trees and street lamps in silhouette against a pink and violet dusk sky, an aircraft climbing away — **Pl. 06**Street lamps, one tree, one plane leaving. The sky did the rest.

A coral and pink sky over a wide tidal flat with hundreds of people standing in the shallow water, mountains behind — **Pl. 03**Pink hour, low tide, and half the city out for it.

Low tide at sunset: an orange and red sky mirrored in the wet sand, a kayak and paddlers in silhouette — **Pl. 01**Low tide, and the whole sky in the water.

Six of thirty. The rest – Kinabalu, the safari, the sea in every mood it has – are on the Photos page. Click any frame to open it full size.

I think a wide field of view matters – at work and everywhere else. Being open to new ideas and new approaches is what lets you reach the answer nobody else reached for, and it is how you grow as a person. Which is why I keep pushing at my own skills, professionally and otherwise.

Lev Sklyarenko by the bay — Kota Kinabalu. Away from the desk I am usually somewhere near water, thinking about something other than day counts.

Thirty frames

Photos

Mostly the sky, mostly at the wrong hour of the day.

Two trips so far: Kota Kinabalu, and Zanzibar with the mainland attached to it. Almost everything here was taken within half an hour either side of sunset, which is either a preference or a habit – I have stopped trying to tell the difference.

I am not trying to document a place. I am trying to catch the moment where it carries something, and then stop pressing the button. Click any frame to open it full size; the arrow keys move through the whole set from there.

The granite crown of Mount Kinabalu under cloud, seen over dark green foliage — **Pl. 05**Mount Kinabalu, with the cloud coming over the top.

Looking up the trunks of very tall rainforest trees towards a bright canopy — **Pl. 07**Straight up, in the rainforest.

An orange and blue sunset over a bay with two islands, people and parked cars along the waterfront road — **Pl. 08**Everyone on the waterfront, waiting for the same thing.

Layers of hills and haze at dusk with a single cloud lit from below — **Pl. 09**Hills, haze, and one lit cloud.

A deep red and pink sunset over water, framed by palm trees in silhouette — **Pl. 10**A sky going properly red.

Looking straight up the trunk of an enormous tree, other trunks radiating around it against the sky — **Pl. 11**Looking up something very old.

A green pasture below a jungle ridge with cloud sitting on the peaks — **Pl. 12**Grass, ridge, cloud, in that order.

Grey granite cliffs under cloud, seen through dark foliage — **Pl. 13**Granite, and the weather that lives on it.

A wide white sand beach with very shallow clear water and a thatched hut on stilts — **Pl. 14**White sand, shallow water, a hut on stilts.

A dark storm front over a green sea, boulders on the sand in the foreground, an island on the horizon — **Pl. 15**Weather arriving, and nobody moving.

Pale overcast light over a calm sea with a low line of surf running up the sand — **Pl. 17**Flat light, low surf.

An overcast sea and sky in near-identical greys, the horizon barely separating them — **Pl. 18**Grey on grey, which I like more than it sounds.

Savanna with flat-topped acacia trees and grazing animals in the middle distance — **Pl. 19**Acacia, grass, and the herd that lives in it.

Seen over the bonnet of a safari vehicle: an elephant standing on the dirt track ahead — **Pl. 20**An elephant deciding whether to move.

A giraffe reaching into the treeline, head and neck above the green — **Pl. 21**A giraffe, mid-chew.

An elephant by a waterhole in thick morning mist, trees dissolving behind it — **Pl. 22**Morning mist, and the one animal in it.

The nave of a church: marble columns, wooden pews and a striped stone floor leading to a lit altar — **Pl. 23**Cool stone, and nobody in a hurry.

A dark wall painted with a list in white capitals, from make it great to make it a legacy — **Pl. 24**A wall with instructions on it.

From an aircraft window: a wingtip above an unbroken floor of cloud, a pale band of sky at the horizon — **Pl. 25**Over the cloud floor, somewhere in the middle.

A jet wing above a cloud deck with a thin orange band of dawn along the horizon — **Pl. 26**Dawn from the wing.

A jet wing over towering white cumulus under a deep blue sky — **Pl. 27**Cloud tops, late morning.

Light aircraft parked on a wet apron at dusk under a heavy storm sky — **Pl. 28**Small planes waiting out the weather.

A star field over a dark treeline, the Milky Way faintly visible — **Pl. 29**The sky over the treeline, no moon.

A dense star field with the faint band of the Milky Way across it — **Pl. 30**Stars, and the Milky Way if you look for it.

Two notes. The last two frames are night skies – they were nearly black as shot, so the levels are lifted enough to make the stars visible. Everything else is the frame as it came off the phone, resized and nothing else.

On in the background

Music

Nineteen records, shelved by whoever made them.

Soul, britpop, rock. Roughly in that order, and it depends entirely on the hour.

Sorted by artist rather than by genre, because a shelf never sorts itself that way. Four Pink Floyd records and three by the Cure is not an accident.

Shelf 01 Pink Floyd 4 records

Pink Floyd, The Piper at the Gates of Dawn: album cover — **The Piper at the Gates of Dawn**1967

Pink Floyd, The Dark Side of the Moon: album cover — **The Dark Side of the Moon**1973

Pink Floyd, Wish You Were Here: album cover — **Wish You Were Here**1975

Pink Floyd, The Final Cut: album cover — **The Final Cut**1983

Shelf 02 The Cure 3 records

The Cure, Kiss Me, Kiss Me, Kiss Me: album cover — **Kiss Me, Kiss Me, Kiss Me**1987

The Cure, Wish: album cover — **Wish**1992

The Cure, Greatest Hits: album cover — **Greatest Hits**2001

Shelf 03 The Beatles 2 records

The Beatles, Abbey Road: album cover — **Abbey Road**1969

The Beatles, 1967-1970: album cover — **1967–1970**1973

Shelf 04 Duran Duran 2 records

Duran Duran, Liberty: album cover — **Liberty**1990

Duran Duran, Duran Duran - The Wedding Album: album cover — **Duran Duran · The Wedding Album**1993

Shelf 05 Kendrick Lamar 2 records

Kendrick Lamar, To Pimp a Butterfly: album cover — **To Pimp a Butterfly**2015

Kendrick Lamar, untitled unmastered.: album cover — **untitled unmastered.**2016

Shelf 06 Oasis 2 records

Oasis, Definitely Maybe: album cover — **Definitely Maybe**1994

Oasis, (What's the Story) Morning Glory?: album cover — **(What’s the Story) Morning Glory?**1995

Shelf 07 Sade 2 records

Sade, Love Deluxe: album cover — **Love Deluxe**1992

Sade, The Best of Sade: album cover — **The Best of Sade**1994

Shelf 08 One apiece 2 records

The Verve, Urban Hymns: album cover — **The Verve · Urban Hymns**1997

ABBA, The Essential Collection: album cover — **ABBA · The Essential Collection**2012

Sleeve art belongs to the artists and labels who made it, and is here at thumbnail size to show a shelf – nothing more.

Get in touch

Contact

Open to new roles.

Analyst roles on a capital markets desk or in a trading-systems team. If a product has to go from a desk requirement into production and then be owned afterwards, that is the work I already do.

Based: Moscow · UTC+3
Relocation: Open to it
Right now: Employed, and happy to talk
Languages: Russian native · English C1 · French B1 · Chinese B1 · Japanese A2

Telegram @levsklyarenko fastest reply Gmail levstudent18@gmail.com Phone +7 963 723 43-02 LinkedIn lev-sklyarenko Instagram @lev__a_