Senior Business Analyst

Senior Business Analyst

Aumega at a Glance

Aumega is a purposebuilt platform that empowers venture capital, private equity, and investment banking firms with realtime financial intelligence. By automating the extraction and standardization of financial data, Aumega enables analysts and decisionmakers to

focus on insights, not hygiene.

Aumega is a Surya Group company, created in 2025.

The Surya Group was founded in 1999 by Mr. D.N. Prahlad (an early employee of Infosys), to create software products. Headquartered in Bangalore, we also have offices in the USA, and France. Other Surya Group companies include Surya Fintech (https://www.surya-fintech.com/), Surya Digital (https://surya-digital.com/), and Avenza Consulting (https://www.avenza-consulting.com/).

What is Needed for the Role

Domain Expertise in IB / PE / VC / AM

• Strong grasp of how investment banks, PE funds, VC funds, and AM firms track portfolio performance over time.
• Hands-on experience reading and interpreting financial reports: P&L, Balance Sheet, Cash Flow, AUM bridges, KPI dashboards, management commentary, segment notes.
• Familiarity with portfolio monitoring packs, LP reporting, covenant tracking, board packs, valuation memos, and credit/investment committee decks.
• Deep understanding of common financial metrics and ratios (revenue, EBITDA, margins, leverage, coverage, cash conversion, AUM flows, IRR/MOIC, churn, LTV/CAC, runway, etc.) and how different investors prioritize them.
• Strong skills in using MS Excel/MS Advanced Excel. Experience in Financial Modelling and metric tracking using MS excel is a bonus and would be prioritized.

Structured Thinking & Schema Design

• Strong experience in designing data dictionaries, and metric catalogs for financial reporting and analytics.
• Able to normalize messy real-world reporting into a stable, extensible schema that works across multiple portfolio companies and industries.
• Comfortable designing hierarchies: metric groups (Revenue, Profitability, Cash, AUM, Insurance KPIs, SaaS KPIs, etc.), dimensions (company, segment, currency, period, scenario), and grain (annual, quarterly, monthly).

Requirement Definition

• Skilled at converting vague stakeholder needs into precise BRDs/PRDs, user stories, acceptance criteria and test scenarios.
• Ability to specify exact extraction rules, edge-cases and what to do when data is missing/inconsistent.
• Strong written communication: you can write long-form specs that engineers and designers can depend on.

Analytical & Quality Orientation

• Obsession with accuracy, auditability, and repeatability of extracted metrics.
• Habit of validating numbers using cross-checks (e.g., reconciliation between statements, ratio sanity checks, trend analysis).
• Comfort designing exception workflows: flags, confidence scores, requires analystreview pipelines.

Stakeholder Management

Able to work closely with:

• Internal engineering teams (backend, ML/LLM, data).
• Product & design teams (to shape flows that match analyst mental models).
• External client stakeholders (Heads of Portfolio Monitoring, Deal Teams, CFO office, Risk, LP reporting).

Comfortable running workshops, requirement sessions and product walkthroughswith senior finance stakeholders.

Nice-to-Have Technical Skills (Bonus but Not Mandatory)

• Ability to design precise, context-rich prompts for LLMs that consistently produce accurate, structured financial outputs

  including clearly defining the analyst persona, specifying document reading behavior (e.g., read full report, handle footnotes/ appendices), instructing how to locate and interpret key financial metrics (revenue, EBITDA, AUM, cash flow, leverage, SaaS KPIs, etc.), enforcing standardized output formats, and explicitly handling edge cases such as missing data, adjusted vs reported figures, segment changes, and rounding or scaling differences.
• Working knowledge of SQL (for validating schemas and checking extracted data).

What You Will Work On

Client & Domain Discovery

Understand:

• How they currently track portfolio performance.
• Which metrics they consider core vs deal-specific.
• How they organize and store portfolio company reports today.
• Map end-to-end workflows: from report download manual extraction Excelmodels / portfolio dashboards management reporting.

Metric Catalog & Taxonomy Design

• Build and maintain a master catalog of financial and operational metrics across multiple asset classes and sectors.
• Define each metric with:

- Clear definition (formula, units, sign convention).

- Typical locations in reports (financial statements, notes, MD&A, KPI tables, footnotes).

- Synonyms and variations used by different companies/industries.

• Classify metrics by:

-Use-case (portfolio monitoring, covenant tracking, valuation, risk, ESG).

-Frequency (monthly/quarterly/annual).

-Materiality (priority metrics vs nice-to-have).

DB Schema & Data Model Design

• Design and iterate on the core database schema that stores:

• Companies, instruments, funds, sectors, regions.

-Financial metrics (fact tables) with period, currency, unit, and scenario dimensions.

-Document lineage (which PDF/page/table a metric came from).

• Work with data engineers to:

-Ensure schemas are normalized enough for LLM extraction and downstream analytics.

-Allow for custom client-specific metrics without breaking the core model.

Prompt & Instruction Design for LLMs

• Write and refine prompt templates and instruction sets that tell the LLM:

-How to parse and read full reports (cover to appendices).

-How to handle synonyms, formatting quirks, footnotes, and charts.

-How to output structured, machine-readable data (tables, JSON).

-What to do when information is missing, ambiguous, or conflicting.

• Design persona-level instructions (e.g., act as a PE portfolio monitoring analyst vs act as a sell-side coverage analyst) where appropriate.

Extraction Evaluation & Ground-Truth Creation

• Lead the creation of gold-standard labeled datasets:

-Select representative sample of company reports (by region, sector, format).

-Manually extract and validate metrics to serve as ground truth.

• Define evaluation frameworks:

-Accuracy / recall / precision of extracted metrics.

-Tolerance thresholds (e.g., rounding, scaling, currency).

-Business-impact KPIs (e.g., time saved per company, reduction in errors).

• Work closely with ML/LLM - developers/engineers to:

-Run experiments across different models / prompts / retrieval strategies.

-Interpret results from a business and domain perspective, not just numbers.

Product Requirements & User Experience

• Collaborate with product managers and designers to:

-Define analyst-friendly flows for uploading PDFs, configuring metrics, and reviewing outputs.

-Ensure the UI surfaces context and traceability (source page, snippet,

confidence level) for each extracted metric.

-Design workflows for human-in-the-loop validation: bulk approval, exception handling, override rules.

• Provide domain input

  on UX micro-decisions (e.g., how to show AUM bridges, currency/unit switches, time-series comparisons).

Pilot Implementations & Client Roll-outs

• Work closely with early-adopter clients to:

-Configure their specific metric sets and templates (e.g., internal portfolio monitoring packs).

-Onboard their existing portfolio companies and historical reports.

-Troubleshoot extraction issues and identify pattern-level fixes.

• Capture feedback systematically

  and translate into product backlog items.

Governance, Controls & Auditability

• Define and document:

-Data provenance rules (which document, which page, which section).

-Confidence scoring and flagging criteria for risky extractions.

-Versioning of prompts, schemas, and metric definitions over time.

• Partner with compliance and security teams to ensure:

-Sensitive data handling aligns with client expectations.

-Multi-tenant / client-specific configurations are safe and isolated.

Thought Leadership & Internal Enablement

• Act as the internal domain champion for all things IB/PE/VC/AM.
• Create playbooks, internal training decks and wiki documentation on:

- How to read and interpret various types of financial reports.

- Common pitfalls in financial metric extraction (e.g., adjusted vs reported, pro-forma, restructuring noise).

- Best practices for domain-driven prompt design.

Your Level of Experience

At Aumega we do not evaluate people based on the number of years of experience they have. Instead, we look for skills that have been gained, regardless of how long it has taken a person to gain those skills.

However, as a point of reference, historically, we have successfully recruited candidates for this role with 3-5 years of prior experience.

Educational & Professional Credentials (Priority)

• Strong academic background in Finance, Accounting, Economics, or related field.
• MBA (Finance), CA/CPA, CFA, FRM or equivalent professional qualification is a strong plus.
• Evidence of continuous learning in data / analytics / AI (courses, certifications, projects).

Skills & Competencies

• Proven ability to read complex financial and regulatory reports and tell the story behind the numbers.
• Hands-on experience building or reviewing detailed financial models (3-statement,DCF, LBO, operating models).
• Comfort working with large volumes of PDFs and unstructured documents.
• Experience writing functional specifications / BRDs that engineering teams have successfully implemented.
• Exposure to or curiosity about:
• Gen AI tools (ChatGPT, Claude, etc.), RAG systems, and prompt engineering.
• Data pipelines and APIs, even if you are not the one coding them.

Prior experience building or using (Bonus)

• Portfolio monitoring systems.
• Credit / investment committee models.
• LP reporting templates and dashboards.

Evaluation Process

The evaluation process starts with a short phone conversation.

After that, the candidate will be invited for a series of interview rounds at our office, which will

be shared with them after the phone screen.

Please note that the Interviews typically take 1-3 hours but may last up to 6 hours.