Siddharth
Jain.
Built production AI across manufacturing, healthcare, and research. Now leaping into Finance × AI — MSc at Frankfurt School, CFA Level 1 in progress, building agentic systems for fintech and banking.
RECRUITER MODE
Too long to read?
Skip the scroll. Ask my AI anything — get the whole story in 60 seconds.
01 / About
More than a
resume.
I'm a Full Stack AI Engineer who's led enterprise AI transformation at CEO level, conducted medical-imaging research at the Coulter BME department (Georgia Tech × Emory), and done cybersecurity work at Georgia Tech — all before finishing my master's degree.
Off the screen, I'm a national-level basketball player, a geopolitics obsessive, an active debater, and a die-hard Manchester United supporter. I think the best engineers are curious about everything — not just code.
Life in frames
Georgia Tech, Atlanta
with the squad
good times
Kashmir, India
out exploring
adventure mode
family first
somewhere in the world
02 / Skills
Four domains. One stack.
Pick a role — see what I ship, what I ship it with, and where it overlaps with the others.
01 / 04
Data Engineer
Pipelines & lakehouses at scale.
50+ tables fed downstream
What I ship
- 01Production pipelines moving TB-scale data daily
- 02Bronze → Silver → Gold lakehouse layers
- 03Real-time ingestion with Kafka + Spark Streaming
Stack
Overlaps with
Tap a role above — or click an overlap to jump
FOUNDATION
In every single project.
These seven carry everything above — whichever role I step into, whichever team I join.
Python
SQL
AWS
Azure
Docker
Git · CI/CD
Linux
Frameworks & tools in daily use
03 / Projects
Things I've Built
Live systems you can try yourself — each with a running demo.
CADUCEUS — Virtual Molecular Tumor Board
Give it a full oncology case — pathology, radiology, genomics, labs — and 55+ specialist agents across 7 layers deliberate in parallel (map-reduce fan-out pushes genuine concurrency past 40, ~150 at peak) to produce a fully-cited, guideline-concordant treatment recommendation with documented dissent, a provenance audit trail, and human-in-the-loop checkpoints. Runs offline on synthetic cases, or LIVE on a real de-identified TCGA patient — variants annotated against CIViC, drug safety from openFDA, trials from ClinicalTrials.gov, literature from Europe PMC. Every claim traces to a retrievable source; a clinician holds the final gate. You watch the 56-agent board light up live over SSE.
PRAETOR — Agentic M&A Due-Diligence Engine
Pick any company on a 3D world globe (searched live across the global GLEIF registry), and a team of 9 specialist agents — integrity, financials, legal, tax, commercial, HR, cyber, ESG, and a live web pulse — run due diligence in parallel on real public data. They cross-reference each other into one red-flag register, then TRIBUNAL renders a defensible GO / NO-GO / CONDITIONAL verdict. Every claim traces to its source (SEC EDGAR filing, OFAC/GLEIF record, or web article); a human holds the final gate. You watch the agents work — and reference each other — live over SSE.
RegRadar — EU Regulatory-Impact Engine
Agentic system that watches the EU regulatory firehose (EUR-Lex / CELLAR), extracts the concrete obligations from each act, maps them to a bank's systems, ranks them by deadline and risk, and drafts the gap-assessment memo in English and German. Every claim is citation-verified against the live EUR-Lex source — programmatically, with a human-approval gate.
CreditForge — Bank-Grade Credit Risk Platform
End-to-end credit-risk stack built to Basel / IRB methodology on Freddie Mac mortgage data. WoE scorecard plus a LightGBM challenger for PD, joined with LGD and EAD into Expected Loss. Leakage-safe point-in-time targets, out-of-time validation, SHAP reason codes, fairness testing, and drift monitoring. A "Risk Copilot" agent team sits on top of the platform tools.
AEGIS Live — Real-Time Streaming AML
A live surveillance wall that scores real Bitcoin-mempool transactions for money-laundering risk the instant they arrive. Resilient WebSocket ingestion → event bus with backpressure → a sliding-window transaction graph → an ensemble of sanctions screening, LightGBM, and anomaly detection → subgraph-explained alerts with LLM-drafted SARs. Fast-path p95 in single-digit milliseconds. Built solo for $0.
Recoupe — Autonomous Subrogation
Multi-agent platform that reads closed insurance claims, assigns fault by jurisdiction, computes the recoverable amount, drafts demand letters, and works counter-offers. Seven specialized agents over a RAG layer of US negligence law + carrier behaviour. Every decision citation-grounded and auditable.
QUORUM — AI Investment Committee
A simulated investment committee of specialized agents — bull, bear, macro strategist, quant/risk officer, PM, and critic — that argue from real market data across structured debate rounds and converge on a documented allocation, with a human holding the final gate. Every number is computed deterministically in Python (SEC EDGAR, prices, FRED); the LLM only narrates. Paper-only, point-in-time backtested vs SPY.
AEOLUS — Renewable Fleet Operations Brain
A governance-first multi-agent "operations brain" for a wind fleet: it detects a degrading turbine, root-causes it, prices acting now vs later against the live German power market and weather, schedules the cheapest safe maintenance window with an OR-Tools CP-SAT solver, drafts the work order, and routes it for one-click human approval — every action behind a policy gate and an immutable hash-chained audit log. The LLM reasons; OR-Tools does the math.
Dam Rehabilitation Chatbot
AI-powered chatbot for dam condition assessment and rehabilitation planning. Guides engineers through structural inspection, interprets damage data, and recommends maintenance strategies.
More on github.com/sidnov6
04 / Stack
From source systems to intelligent action.
Most data engineers stop at the warehouse. Most ML engineers start there. I own the entire journey — from connecting to enterprise systems to the AI products built on top.
SOURCES
50+ enterprise systems
TRANSFORM
Spark · dbt · Airflow
LAKEHOUSE
Bronze · Silver · Gold
AI / BI
Dashboards · GenAI · ML
SOURCES
SAP ERP
Enterprise Resource Planning
SQL Server
.NET Portals & Internal Apps
Salesforce
CRM & Sales Pipeline
SharePoint
Documents & Files
Workday
HR & Payroll Records
TRANSFORM
Pipeline
ETL · ELT · Real-time + Batch
Spark · dbt · Airflow
LAKEHOUSE
Snowflake · Databricks
AI / BI
BI Dashboards
Power BI · Tableau
GenAI Chatbots
LangChain · RAG · LLMs
ML Models
PyTorch · Scikit-learn
Predictions
Forecasting · Anomaly
ONE PERSON · FULL STACK · END-TO-END OWNERSHIP
Source Integration
Direct connectors to enterprise systems — SAP ERP, SQL Server, Salesforce, SharePoint, Workday. Real-time CDC streams or scheduled batch pulls, whatever the system demands.
Transformation & Modeling
ETL/ELT pipelines built with Airflow, dbt, and Apache Spark. Data quality, lineage, and orchestration baked in. Refresh cadence configurable from real-time streaming to nightly batches.
Lakehouse Architecture
Medallion design on Snowflake or Databricks. Bronze (raw) → Silver (cleaned) → Gold (business-ready). One governed source of truth for the entire organization.
AI & BI Layer
Power BI dashboards, GenAI chatbots, ML models, and predictive analytics — all consuming from the same governed gold layer. End users get answers, not pipelines.
05 / Journey
Suzlon Energy
AI Engineer — CEO Office, Manufacturing
Manufacturing AI, Data & Digitisation Strategy
Jun 2025 – Jun 2026 · Pune, India
Drove AI, BI, and data initiatives across Suzlon's manufacturing organisation — 10 plants in Asia and Europe, 300+ daily users, 14 BI dashboards, 7 production GenAI bots, 250+ KPIs.
Wind Energy
- Delivered 14 enterprise BI dashboards and shipped 7 production GenAI bots (RAG + SQL agents) across Quality, Safety, Production, Energy, and HR — serving 300+ daily users across 10 manufacturing plants in Asia and Europe.
- Worked in a 6-person engineering team to build Suzlon's Manufacturing Control Tower — a consolidated executive view of 250+ KPIs across SQPDCME (Safety, Quality, Productivity, Delivery, Cost, Manpower, Environment).
- Helped rebuild Suzlon's monthly Manufacturing Business Review — sourced and modelled data from 50+ operational systems and deployed agentic AI that answers live questions with on-the-fly charts, cutting manual dashboard and Excel reporting.
- Contributed to Suzlon's AI, Data & Digitisation roadmap — the operating model, tooling stack, and rollout plan.
scroll to explore
05 → The Pivot
Manufacturing was yesterday.
Finance is today.
One year of manufacturing AI at the CEO Office showed me what AI can do at scale. Now — back at Frankfurt School full-time — I am taking that playbook to a $50 trillion industry: finance.
Yesterday
Manufacturing AI
Suzlon Energy · CEO Office
THE LEAP
2026 →
Today
Finance AI
Quant · Fintech · Bank AI
Why now
Finance needs AI — urgently.
AI's contribution to global GDP by 2030
— PwC, Sizing the Prize
Annual GenAI value potential in banking
— McKinsey GenAI Report
Financial-services AI spend by 2027
— IDC, 2024
Finance leaders deploying GenAI
— Accenture / Citi GPS
What the industry says
The most influential voices are aligned.
Jamie Dimon
CEO, JPMorgan Chase
“AI may prove to be as transformational as some of the major technological inventions of the past several hundred years — printing press, electricity, the internet.”
Annual Letter, 2024
Jane Fraser
CEO, Citigroup
“GenAI will turn knowledge workers into super-knowledge workers. It will revolutionize financial services in the next decade.”
Citi GPS Report, 2024
Larry Fink
CEO, BlackRock
“AI has the potential to transform every aspect of our business and the broader financial system — from research, to risk, to client service.”
Davos 2024
Ted Pick
CEO, Morgan Stanley
“Wealth management may well be the most powerful AI use case in our industry. We are only scratching the surface.”
Q1 Earnings Call, 2024
What I want to build
AI × Finance: the use cases I am excited about.
Algorithmic Trading
ML-driven strategies, sentiment models, high-frequency signals
Fraud Detection
Real-time anomaly detection across millions of transactions
Credit Risk ML
Alt-data scoring beyond FICO — fairer, faster, more accurate
Robo-Advisors
Autonomous portfolio construction, rebalancing, tax loss harvest
Document AI
10-Ks, prospectuses, contracts, ISDA agreements — at scale
Agentic Research
AI agents that read filings, news, and build investment theses
RegTech & Compliance
AML, KYC, transaction monitoring — automated, auditable
Earnings NLP
Sentiment + tone analysis on earnings calls and shareholder letters
Bridging the gap
Technical depth meets finance fluency.
To build the right AI agents for finance, you need to speak the language of money. So I am not just leveling up technically — I am pursuing the CFA Level 1.
CFA · LEVEL 1 · PURSUING
Currently studying
2026 sitting · not yet cleared
AI ENGINEERING · LEVELING UP
Deeper in the stack
Agentic + RAG systems
AI agents that actually understand finance.
How you can help
I’m looking for people who
want to take this leap with me.
If you build, hire, mentor, or invest in Finance + AI — or know someone who does — please reach out.
Roles in fintech, quant, or bank AI teams
Mentors with finance + ML backgrounds
Collaborators on agentic finance projects
Intros to teams shipping AI in production
06 / Education
Academic Foundations
From VIT Vellore to Frankfurt School — building the foundations that power real-world enterprise AI.
FT #32 (Finance & Mgmt) · Frankfurt
MSc Artificial Intelligence & Data Science
Frankfurt School of Finance & Management
2026 – 2028 · Frankfurt, Germany 🇩🇪
Pursuing an MSc in AI & Data Science at one of Europe's leading finance and management schools, located in the heart of Frankfurt — Germany's financial capital.
- Ranked #32 worldwide — among the strongest global positions in Finance & Management (FT Global Rankings)
- Specialising in applied AI for financial and industrial systems
- Located in Frankfurt — the financial capital of continental Europe
- Research focus: enterprise LLMs, agentic finance, and data-intensive AI systems
Key Courses
#12 India · Vellore
B.Tech — Information Technology
VIT Vellore
2021 – 2025 · Vellore, Tamil Nadu, India 🇮🇳
Completed B.Tech in IT from one of India's top-ranked private engineering universities, while completing 3 international research internships across the USA and India.
- Ranked #12 in India among top universities (NIRF + QS rankings)
- Completed 3 international research internships during undergrad — Georgia Tech, Coulter BME (GT × Emory), IIT Jammu
- National-level basketball player throughout all 4 years
- Led the ACM Student Chapter as Operations & Marketing Head — raised $11K in sponsorships
Key Courses
07 / Life
Beyond the Terminal
What happens when the terminal is closed.
National Level Basketball
Basketball isn't a hobby — it's where I learned everything that matters: reading situations in real time, leading under pressure, trusting teammates. Playing at national level shaped how I think about systems, strategy, and performing when it counts.
Manchester United
Die-hard Red Devil through every trophy and every rebuild. Supporting United taught me loyalty, resilience through brutal seasons, and absolute belief in systems even when short-term results disagree. Glory Glory Man United — we always come back.
Geopolitics & World Affairs
I follow geopolitics the way some people follow football — obsessively, analytically, with strong opinions. From Indo-Pacific power shifts to EU economic architecture, from energy markets to defense policy. Understanding power structures makes me a sharper systems architect.
Debate & Oratory
Debating sharpens the mind in ways no technical course can. I participated in collegiate competitions and Model UN, and I believe deeply in structured argumentation and evidence-based reasoning. It's made me a sharper thinker and a much better presenter to CXO stakeholders.
07 → Giving Back & Learning
Teach what I know. Learn what I need.
Taught 250+ students, raised $11,000 as a student ops lead — and every day, the next skill I need.
Giving Back
Volunteer Educator
Becoming I Foundation
Mar 2022 – Aug 2024 · Vellore, India
Python & Mathematics — 4 Government Schools
- Taught Python programming and Mathematics to ~200 students across 4 government schools in Tamil Nadu.
- Designed a beginner-friendly curriculum for students with no prior computer exposure.
- Built long-term mentoring habits and public-speaking confidence through grassroots teaching.
Volunteer Educator
Suzlon Energy CSR
Jun 2025 – Jun 2026 · Pune, India
AI Literacy & Digital Inclusion
- Ran weekly AI literacy sessions for children of plant staff and underserved communities.
- Taught foundational AI, digital literacy, and logical problem-solving to 50+ students.
- Translated complex ML concepts into accessible, beginner-friendly modules.
Operations & Marketing Head
ACM Student Chapter — VIT
Mar 2022 – Aug 2024 · VIT Vellore
Association for Computing Machinery
- Raised ~$11,000 in sponsorships through 200+ cold calls and corporate outreach.
- Ran end-to-end ops for multi-day events with 500+ participants — hackathons, ideathons, talks.
- Led cross-functional student teams across operations, marketing, design, and tech.
Always Learning
Agentic AI Systems
Daily · always shipping
Building agentic finance projects in LangGraph, MCP servers, RAG pipelines over financial documents, and LLM evaluation harnesses. Treat learning as a build practice, not a course list.
Reading & Research
Capital markets · geopolitics · AI papers
Following the agentic finance shift across JPMorgan, Citi, Goldman, Two Sigma. Reading earnings letters, CEO letters, and frontier AI research weekly. I write the best of it up on the blog.
"If I am not learning, I am not building. If I am not teaching, I am not learning."
— personal mantra
08 / Writing
Notes on AI in Finance.
Where the agentic moment meets capital markets — what is happening, why it matters, and what I am building toward.
09 / Contact
Let's Connect
AI opportunity, research collaboration, or you just want to debate geopolitics — I'm all ears.
Actively looking for full-time AI/ML engineering roles, research collaborations, and interesting problems to solve.