Daniil Daletski

Shift Operational Manager

Studioworks OÜ | Tallinn, Estonia | Current role: 2025-present; company tenure: 2021-present

Own shift-level operations, workforce planning, monthly reporting and operational analysis across two portfolios and up to 19 service lines. Scope includes up to 61 employees concurrently scheduled and an HR-sourced workforce planning base of approximately 97 FTE.

• Capacity planning. Translate demand into service-hour and FTE requirements, balancing staffing supply, skill certification, rotation constraints and planned closures. Planned up to 10,900 service hours per month and 49,800 across the January-May 2026 cycles; reconcile requirements by service line.
• Availability. Delivered 99.84% combined service availability across 49,900 agreed hours in January-May 2026, meeting the contractual KPI with 81 hours of unplanned downtime. Attribute downtime by driver and distinguish frequent incidents from those with the greatest operational impact.
• Incident governance. Own a register of 170 incidents across 11 monthly cycles, including 168 with timed start-to-fix durations. Rebuilt resolution-time reporting from source records and cleared 48 pre-existing formula errors.
• Reporting & data quality. Own a 10-sheet monthly model covering two portfolios, 10+ KPIs and five source feeds, with 3,400+ formula cells and eight defined KPI steps. Validated 140,000+ records over five cycles, resolving duplicates, voided entries and cross-sheet inconsistencies before publication.
• Assumption testing. Identified systematic throughput and average-handling-time assumption drift across five consecutive cycles. Re-based service-hour requirements, quantified effective staffed hours against nominal headcount and escalated capacity risks to Scheduling.
• People leadership. Run performance evaluation against an anchored 0-5 scale with defined observation minimums and 10 assessments per employee per month; reconcile scoring across evaluators. Lead coaching, feedback and structured improvement plans. Interview candidates and contribute directly to hiring decisions. Communicate assumptions, constraints and trade-offs to Scheduling, HR, IT and operational leadership.

Career progression: front-line operations from 2021; senior operational responsibility from 2023, covering quality evaluation, incident handling, escalation and shift coordination; Shift Operational Manager from 2025.

• Detailed operations record ↗

Sales Representative

Southwestern Advantage | Texas, USA | Summer 2024

Ran an independent sales territory through a 12-week field programme conducted entirely in English. Planned daily activity from conversion data, adapted targets weekly and worked autonomously in an international environment.

Employee Performance Platform

Company-wide internal product | In production | Python, SQL, web application

Self-initiated platform giving 136 employees personal real-time dashboards and replacing manual manager lookup-and-relay with direct access to evaluations. Carries 140-280 evaluations per week.

• Unified evaluations, errors, attendance, attitude and skills in a database-backed system with role-based access, HR attendance integration and a live error feed.
• Built a weekly leadership digest covering volume, score changes, top movers, a 14-day not-evaluated coverage list, multi-month decline alerts and a narrative of the week's priorities.
• Connected feedback to a written six-category weighted framework covering 21 incident types; management remains responsible for coverage follow-through and scoring consistency.

• Platform case study ↗

Daily operational performance evaluation tool

Deployed full-stack product | In production | Reviewed reporting and scheduled delivery

Transforms batches of up to 10 evaluation screenshots into traceable records, reviewed analytics and monthly workbooks under controlled access.

• Retains the source screenshot against each record. Extraction produces candidates; uncertain or unresolved rows remain flagged for a two-phase human review gate.
• Feeds approved records into analytics and conditional workbook output, followed by scheduled monthly delivery. Human review remains part of the workflow; automatic extraction accuracy is not assumed.

• Live product ↗
• Public source ↗

Operational performance model & quality tooling

Internal operational systems | Advanced Excel, SQL, HTML and JavaScript

Built the monthly performance and availability calculation layer: 10 sheets, approximately 2,900 calculation rows and eight steps from possible hours through FTE and line coverage to error rates and mean resolution time. Reconciled against five source feeds with plan-versus-actual commentary.

• Developed the six-category quality evaluation framework, a 0-5 anchored scoring scale, an offline scenario trainer and a monthly schedule-review matrix. Govern a 55-code error taxonomy covering 450+ coded exception records and 190 voided transactions.

NEXUS Terminal

Native research workspace | Research and paper state | TypeScript, Rust, Tauri, React

Built a decision workspace with 96 registered views, 67 native Tauri commands and six evidence layers. Integrates source evidence, model results and risk checks rather than separating the output from its assumptions.

• Architecture spans external data, BigQuery / Cloud Run / Pub/Sub processing, local Rust computation and a cross-cutting audit layer for warnings, decision gates and paper-state outcomes.

Research and paper state only. No live trading performance or production execution-control readiness is claimed.

• System case study ↗
• Public research branch ↗

Quant Lab

Interactive educational instruments | TypeScript, Canvas and SVG

Built four browser-based instruments for Monte Carlo simulation, portfolio optimisation, stress analysis and option pricing. Supports up to 20,000 Monte Carlo paths and a 4,000-portfolio cloud with seeded, reproducible calculations and analytical checks alongside results.

Models include geometric Brownian motion, a calm/crisis regime chain, the Markowitz frontier, parametric VaR / Expected Shortfall and Black-Scholes with delta hedging. Educational user-input scenarios; no live market feed or calibration to real assets.

• Open the instruments ↗

Online Information Flow and Short-Term Stock Returns

Tested whether abnormal search activity, news intensity and financial-news sentiment add information to short-horizon equity-return models beyond market, volatility, liquidity, size, value, momentum and sector effects.

• Dataset. 25,876 firm-week observations across 99 large-cap firms (50 U.S., 49 European), covering 2021-2025. Engineered 13 model-ready variables from five sources, with approximately 130,000 outbound collection calls.
• Methods. Benchmarked pooled OLS with firm- and week-level double-clustered standard errors against XGBoost, Random Forest and Elastic Net. Used a 243-setting search grid and approximately 2,916 fits; assessed model differences with Diebold-Mariano tests and interpreted predictions with Tree SHAP.
• Validation. Trained on 2021-2023, tuned on 2024, froze preprocessing and model selection, then evaluated once on the untouched 2025 window. Fixed seeds, cached inputs, version pinning and stage-level audit logs support a 19-module Python pipeline and an approximately 70-minute clean-clone reproduction.

Estonian Business School | 2023-2026

International Business Administration | Cum laude | Finance focus

Completed the degree alongside full-time operational work. CFA-aligned risk coursework covered Value at Risk, Expected Shortfall, stress and scenario analysis, Monte Carlo methods, and model assumptions and limitations.