Skip to content
View LalitSatvik's full-sized avatar
😶
existing
😶
existing

Block or report LalitSatvik

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
LalitSatvik/README.md

Lalit Satvik Surisetty

Engineer working across decision systems, machine learning, and full-stack product — Hyderabad, India. Fraud and risk analytics, agentic LLM workflows, data pipelines, and the web apps that sit on top of them.

Portfolio · LinkedIn · Email

Currently open to engineering roles. Most of what I've been building lately is below.

Selected projects

  • agentic-fraud-investigation-system — a classical fraud-scoring model paired with an LLM investigation agent: enrichment tools, a human review queue, an audit log, and an evaluation harness that scores the agent against labelled cases.
  • doc-unify — offline-first document intelligence. Ingests mixed formats, proposes a unified schema, then extracts against it with provenance back to the source span.
  • federated-learning-portfolio — FedAvg vs FedProx on the LEAF benchmark, simulated with Flower and streamed to the browser round by round. Includes an honestly-reported negative result.
  • open-data-job-board-aggregator — a searchable board built from free public job APIs, with a personal kanban application tracker and CSV/JSON export.
  • tdacn — temporal drift analysis of XBRL accounting-concept networks using node2vec graph embeddings, with an interactive D3 dashboard.

Skills

Experience

Junior Risk Analyst — CITIC Middle East Contracting, Dubai · Jun 2025 – Jul 2026

  • Maintained ~800 risk-register entries across 5 active projects on Amazon RDS with controlled access
  • Reviewed 80+ FIDIC-based contracts, flagging liability, indemnity, and penalty exposure
  • Built multi-currency exposure tracking and variance reporting on AWS Lambda; published findings through Amazon QuickSight
  • Automated recurring reporting workflows, cutting manual analysis time by roughly 60%

BI & Analytics Intern — Innodatatics, Hyderabad · Dec 2024 – Mar 2025

  • Delivered 30 Power BI and Tableau dashboards across business, risk, and ESG domains
  • Built an ESG Risk & Sustainability Intelligence Platform integrating 20–30 material metrics
  • Cleaned, transformed, and validated 8 core business datasets with Python, SQL, and pandas

Education

B.Tech, Computer Science Engineering — SRM University, Chennai · 2022 – 2026 · CGPA 8.53/10

Machine Learning Specialization — DeepLearning.AI / Stanford Online (Coursera)

Pinned Loading

  1. agentic-fraud-investigation-system agentic-fraud-investigation-system Public

    Classical risk model + LLM investigation agent with tools, human review queue, and evaluation harness

    Python 1

  2. doc-unify doc-unify Public

    Offline-first document intelligence — multi-format ingest, unified schema proposal, extraction with provenance

    Python 1

  3. federated-learning-portfolio federated-learning-portfolio Public

    FedAvg vs FedProx on LEAF, simulated with Flower, streamed live

    Python 1

  4. tdacn tdacn Public

    Temporal drift analysis of XBRL accounting concept networks via graph embeddings

    Python 1

  5. invoice-payment-reconciliation invoice-payment-reconciliation Public

    Invoice-to-payment reconciliation: upload invoices + bank statements, deterministic matching engine, review UI, export.

    Python 1

  6. open-data-job-board-aggregator open-data-job-board-aggregator Public

    Job board aggregator with a personal application tracker

    Python 1