Skip to content
Open to remote · Mumbai, India

Sharath Dinesh Data Analyst & Data Automation Expert

I turn enterprise-scale financial data into automated, trustworthy systems.

view_work get_in_touch
4.3+
yrs in data
40
enterprise clients
99%+
data accuracy
~50h
saved / week
~/sharath — pipeline.py

$ python pipeline.py --client enterprise

[ok] connecting postgres ......... done

[ok] ingesting 5 sources via n8n ... done

[ok] control totals validated ...... 100%

[!!] anomalies flagged .............. 3

[ok] reconciliation accuracy ....... 99.4%

$

currently

Data Analyst Manager @ Cointab Software Pvt Ltd

01 / about

Who I am

Data analyst by title, automation engineer in practice.

Data Analyst / Automation Engineer with 4+ years turning enterprise-scale financial data into automated, trustworthy systems. I own data quality and root-cause analysis across reconciliation datasets (balance sheets, income statements, MIS) at 99%+ accuracy, and translate business requirements into technical solutions across concurrent client engagements. I build reusable Python and SQL workflows (n8n, FastAPI, Playwright) that replace manual effort, and report findings through BI dashboards (Power BI, Metabase) for cross-functional stakeholders. Google AI Essentials certified.

Python SQL n8n PostgreSQL Metabase Power BI
01

Data Quality

Root-cause analysis across reconciliation datasets at 99%+ accuracy.

02

Automation

Reusable Python + n8n + Playwright pipelines replacing manual effort.

03

Reporting

PostgreSQL modelling with Power BI and Metabase for stakeholders.

04

Business Analysis

BRDs and requirement gathering across concurrent client projects.

02 / experience

Where I've worked

Four years of compounding ownership at Cointab, plus published academic research.

  1. Data Analyst Manager

    Cointab Software Pvt Ltd · Malad, Mumbai

    Apr 2024 — Present

    2 yrs 4 mo

    Leading data quality, automation and reporting for enterprise financial reconciliation across 40 clients.

    • Own data quality and root-cause analysis across financial reconciliation data (balance sheets, income statements, MIS) for 40 enterprise clients — traced anomalies to source and fixed underlying processes, sustaining 99%+ accuracy via SQL and automated validation checks.
    • Gather and document business requirements (BRD), translating them into technical solutions across 3–5 concurrent client projects with cross-functional teams.
    • Built reusable, scalable Python + n8n workflows automating reconciliation and reporting pipelines — replaced one-off scripts, cut manual processing time 35%.
    • Automated performance/accuracy review of API data ingestion via n8n feeding Cointab Reconciliation Software — cut manual data-pull time 70–80%.
    • Automated end-to-end data extraction across 5–6 external sources using Playwright — saved ~50 hrs/week.
    • Own the internal dashboard and reporting layer (PostgreSQL + Metabase) for 5–6 cross-functional stakeholders.
    • Built KPI dashboards on traffic and engagement for 3 new products — surfaced patterns that informed UI and retention decisions.
    • Manage and train a team of 5–6 data analysts on SQL, automation best practices and data-quality frameworks — cut error rates 90%.
    PythonSQLPostgreSQLn8nPlaywrightMetabaseFastAPIPandas

  2. Senior Data Analyst

    Cointab Software Pvt Ltd · Malad, Mumbai

    Jun 2023 — Apr 2024

    10 mo

    Scaled reconciliation analysis and automation across a growing enterprise client base.

    • Led reconciliation analysis for high-volume enterprise clients, formalising validation checks into repeatable SQL suites.
    • Introduced n8n-based automation into the reporting pipeline, removing recurring manual pulls.
    • Mentored junior analysts on SQL performance and data-quality practices.
    SQLPythonn8nMetabase

  3. Data Analyst

    Cointab Software Pvt Ltd · Malad, Mumbai

    May 2022 — Jun 2023

    1 yr 1 mo

    Built the analysis foundation for client reconciliation datasets and internal reporting.

    • Performed daily reconciliation of financial datasets and root-caused mismatches to source systems.
    • Automated recurring Excel and SQL reporting into scripted pipelines.
    • Supported BRD documentation and client onboarding for new reconciliation engagements.
    SQLPythonExcelPandas

  4. Academic Research & Projects

    KJ Somaiya College of Engineering · Vidyavihar, Mumbai

    May 2020 — Dec 2020

    7 mo

    Peer-reviewed research applying Python and deep learning to clinical datasets.

    • Published peer-reviewed research (Annals of Data Science, Springer) applying Python-based contextual classification to 3,000+ COVID-19 patient records.
    • Built a deep learning pipeline for ECG signal digitization achieving 94–97% arrhythmia-detection accuracy across 500 binary images.
    PythonPandasRegexDeep LearningTensorFlow
03 / projects

Systems I've built

Automation and analytics work, framed as problem → solution → impact.

Data Automation

Reconciliation Automation Engine

Reusable Python + n8n workflows that automate financial reconciliation and reporting pipelines for 40 enterprise clients.

problem
Reconciliation reporting relied on one-off scripts per client, making delivery slow and error-prone.
impact
Cut manual processing time 35% and sustained 99%+ data accuracy across enterprise reconciliation datasets.

35%

Manual time cut

99%+

Accuracy

40

Clients

Pythonn8nPostgreSQLPandas
Data Quality

API Ingestion QA Automation

Automated performance and accuracy review of API data ingestion feeding Cointab Reconciliation Software.

problem
Ingestion accuracy was reviewed manually, requiring repeated data pulls and slowing issue detection.
impact
Reduced manual data-pull time 70–80% and shortened detection time for ingestion issues.

70-80%

Data-pull time cut

n8nPythonFastAPISQL
Automation

Playwright Multi-Source Data Extraction

End-to-end automated extraction across 5–6 external sources, eliminating manual collection.

problem
Analysts spent large parts of the week manually collecting data from external portals.
impact
Saved roughly 50 hours per week of analyst effort.

~50

Hours saved / week

5-6

Sources

PlaywrightPythonn8n
Analytics

Internal Data Quality & Finance Dashboards

PostgreSQL + Metabase reporting layer surfacing data-quality and financial metrics for cross-functional stakeholders.

problem
Data-quality and financial metrics lived in ad-hoc queries with no shared view.
impact
Gave 5–6 cross-functional stakeholders a single source of truth for data quality and finance metrics.

5-6

Stakeholders

PostgreSQLMetabaseSQL
Analytics

Product KPI & Engagement Dashboards

Traffic and engagement KPI dashboards for 3 new products, informing UI and retention decisions.

problem
New products launched without visibility into traffic, engagement or retention patterns.
impact
Surfaced usage patterns that directly informed UI and retention decisions for 3 products.

3

Products

Power BIMetabaseSQLPython
Machine Learning

ECG Signal Digitization (Deep Learning)

Deep learning pipeline digitizing paper ECG traces for automated arrhythmia detection — published research.

problem
Paper ECG records could not be analysed programmatically at scale.
impact
Achieved 94–97% arrhythmia-detection accuracy across 500 binary images; published in the Journal of Medical and Biological Engineering.

94-97%

Accuracy

500

Images

PythonDeep LearningComputer VisionNumPy
04 / skills

Toolkit

What I reach for, and how deep I go.

Languages

  • Python Expert
  • SQL (PostgreSQL / MySQL) Expert

Automation & Tools

  • n8n Expert
  • FastAPI Advanced
  • Playwright Intermediate
  • Pandas Expert
  • NumPy Advanced

Visualization & Reporting

  • Power BI Advanced
  • Metabase Expert
  • Excel Expert
  • Matplotlib Advanced
  • Seaborn Advanced

Business Analysis

  • BRD & Requirement Gathering Advanced
  • Stakeholder Management Advanced
  • Cross-functional Collaboration Advanced

Data & AI

  • Data Reconciliation Expert
  • Anomaly Detection Advanced
  • Machine Learning Intermediate
  • Deep Learning Intermediate
  • Google AI Essentials Certified
05 / research

Published research

Peer-reviewed work from my time at KJ Somaiya.

Springer2022

Contextual Classification of COVID-19 Patient Records

Annals of Data Science

Python-based contextual classification (pandas, regex) applied to 3,000+ COVID-19 patient records to structure unlabelled clinical text.

PythonPandasRegexNLP
DOI: 10.1007/s40745-022-00378-9
Springer2021

Deep Learning Pipeline for ECG Signal Digitization

Journal of Medical and Biological Engineering

Deep learning pipeline digitizing paper ECG traces, achieving 94–97% arrhythmia-detection accuracy across 500 binary images.

PythonDeep LearningComputer Vision
DOI: 10.1007/s40846-021-00632-0
06 / education

Education & certifications

B.Tech · Electronics and Telecommunication

KJ Somaiya College of Engineering

Vidyavihar, Maharashtra

Aug 2018 — Jun 2022

  • Published two peer-reviewed papers during undergraduate research.
  • Focus on signal processing, machine learning and data analysis.

Google AI Essentials

Coursera

verify

Google Prompting Essentials Specialization

Google

verify

$ more certifications in progress...

07 / writing

From the blog

Notes on data engineering, automation and analytics.

·5 min

Automating Financial Reconciliation with n8n and Python

A walkthrough of the architecture behind a reconciliation automation layer that cut manual processing time by 35% across 40 enterprise clients.

AutomationData EngineeringPython
read_all_posts →
08 / contact

Let's build something

Automation, analytics or a reconciliation problem that refuses to die — I'm interested.

$ cat contact.json

{

"email": "work.sharathdinesh@gmail.com",

"phone": "+91 96997 78182",

"location": "Mumbai, India",

"linkedin": "/in/sharathdinesh",

"status": "Open to remote"

}

Stored securely in Supabase. No spam, ever.