Experience: 2.00 + years
Salary: Confidential (based on experience)
Expected Notice Period: 15 Days
Shift: (GMT+05:30) Asia/Kolkata (IST)
Opportunity Type: Hybrid ()
Placement Type: Full Time Permanent position(Payroll and Compliance to be managed by: MDMY)
(*Note: This is a requirement for one of Uplers client - MDMY)
What do you need for this opportunity
Must have skills required:
Industrial / Sensor / Iot Data Experience, Mechanical / Chemical / Petroleum Engineering, Multivariate Analysis, Signal Processing Knowledge, Stakeholder experience, Startup / Growth Working Environment, Matplotlib / Plotly, NumPy, pandas, SQL Proficiency, Time-series data experience, Python
MDMY is Looking for:
Data Analyst
Location: New Delhi / Gurugram — Hybrid (2–3 days on-site)
Function: Data Science / AI
Reports to: Data Ops & Analytics Lead
Compensation: Competitive
The Opportunity
Most data analyst roles give you a question and ask you to answer it. This one gives you a dataset and asks you to find the question.
Our clients operate complex industrial equipment — machines that generate continuous streams of sensor data across dozens of channels: speed, pressure, temperature, flow rates, vibration, efficiency. When a client says something doesn't look right, there is no ticket, no predefined dashboard, no SQL query that surfaces the answer. There is data, domain context, and the skills to work through it systematically until
something explains what is happening.
That investigation — forming a hypothesis, pulling the right signals, building the right visualisation, testing and discarding and retesting — is the core of this role.
The clients are enterprise operators making decisions on assets worth hundreds of millions of dollars. When you find a pattern that explains a performance degradation, it saves real money. When your analysis identifies that equipment is operating outside its optimal range, that recommendation goes to an engineer who acts on it. The stakes are higher than a business dashboard, and the work is correspondingly more interesting.
You do not need a domain background to do this job well. The industry context is learnable — especially with the AI tools and platform you will have access to. What you need is the investigation instinct: the ability to approach an undefined problem in a large, complex dataset and work through it systematically until you understand what is happening.
About Us
We build an enterprise AI platform for real-time monitoring, diagnostics, and predictive
maintenance of industrial equipment.
- 45+ employees across New Delhi and Houston
- Venture-funded, closing Series A | 80–90% YoY growth
- 15+ enterprise clients across oil & gas, power generation, and LNG
- Physics-informed ML + production LLM/VLM: a 12–18 month lead over the field
What You'll Own
Data Investigation & Root Cause Exploration (50%)
You take ownership of open-ended analytical questions on client operational data. A machine is behaving strangely and no one knows why. You pull the data, look at the signals, form a hypothesis, test it, discard it, form another — and you do this methodically, not randomly, until you have a defensible explanation and a recommendation.
- Pull and structure raw sensor data from the platform across relevant time windows and channels
- Formulate testable hypotheses about what is driving an observed behaviour or anomaly
- Build exploratory analyses — distributions, correlations, time-series decompositions, multivariate comparisons — to test and refine hypotheses
- Document your investigation approach, findings, and reasoning clearly enough that another analyst can follow and replicate
- Work with domain engineers to pressure-test findings against subject-matter knowledge — you bring the data rigour, they bring the equipment intuition
Performance & Efficiency Analysis (25%)
You answer quantitative questions about how equipment is operating relative to its potential.
- Analyse relationships between operating variables to identify efficiency patterns and anomalies
- Build operating-point analyses — where is this equipment running vs. where it should be, and what does the data say about optimal conditions
- Surface degradation patterns — performance trends that are drifting or narrowing — before they become costly events
- Translate quantitative findings into clear, actionable recommendations for engineering and operations teams
Visualisation & Communication (15%)
You make complex multivariate analysis understandable to people who are not data scientists.
- Build visualisations that communicate relationships in sensor data to an audience of engineers, not analysts
- Present investigation findings clearly — what you looked at, what you found, what it means, what you recommend
- Produce written summaries of analyses that become part of the institutional knowledge base
Analysis Infrastructure & Reusability (10%)
You turn one-off investigations into repeatable capability.
- Document analysis patterns that work — data structures, visualisation approaches, investigation frameworks — so they can be reused across similar problems
- Contribute to the shared knowledge base of equipment behaviour patterns and analytical approaches
- Identify where recurring analyses could be automated or templated into the platform
What Success Looks Like
First 30 days: You have deep familiarity with the data structure — how data is organised across clients and equipment types, and the analytical tools available. You have completed your first independent investigation, even a small one, from data pull to documented finding.
First 60 days: You are independently handling investigation requests with minimal hand-holding. Your analysis outputs are clear, well-reasoned, and useful to the engineering team. You are building fluency in the domain through the data itself.
First 6 months: The team has a reliable, high-quality analytical resource for open-ended data questions. Investigation turnaround is fast. Your analyses are informing real decisions about client equipment. You have started building repeatable frameworks that can be applied across clients.
Your success metric: When an engineer has an open-ended data question about client equipment, sending it to you is the obvious first step — because you consistently come back with a clear, well-structured answer.
Who You Are
Must-Have
- 2+ years doing data analysis where the questions were not predefined — you have had to
formulate the investigation, not just execute a known query. Exploratory, open-ended work is your default, not an exception.
- Strong Python for data analysis — pandas, matplotlib/plotly (or similar visualisation libraries), scipy for statistical testing. Not ML or modelling — analysis and exploration. You should be fast and fluent here.
- Hypothesis-driven approach — when you encounter an anomaly in data, your instinct is to form a specific, testable explanation and work through it systematically. You do not thrash randomly through a dataset hoping to find something.
- Comfortable with ambiguity in the problem — something doesn't look right is a valid starting point for you. You know how to turn an undefined problem into a structured investigation.
- Time-series data experience — you have worked with data that has a temporal dimension, where the order of observations matters and patterns evolve over time.
- SQL proficiency — you can pull and structure data from relational databases without assistance.
Strong Signals
- Experience communicating analytical findings to non-data-scientist audiences (engineers, operations teams, business stakeholders)
- Signal processing basics — filtering, decomposition, frequency analysis on time-series data
- Familiarity with multivariate analysis techniques — PCA, correlation analysis, clustering for pattern discovery
- Engineering degree or coursework (any discipline) — analytical foundation that accelerates ramp-up
- IoT, sensor, or operational data background — working with equipment telemetry, SCADA data, or similar real-world datasets. Not required, but it compresses your ramp significantly.
- Experience at an IoT, energy, or manufacturing company
The Right Mindset
- Investigative by default — you find open-ended data problems more interesting than well-defined ones. The I don't know what's causing this yet state is energising, not frustrating.
- Rigorous, not exhaustive — you know how to build a focused, efficient investigation rather than running every possible analysis. Your hypotheses guide what you look at.
- Domain-curious — you do not need to understand the physics of industrial equipment on day one, but you find it genuinely interesting that the variables you are analysing are physically related to each other. The domain context makes the data more meaningful to you, not less.
- Clear communicator — your analyses produce a finding, not just a notebook. You can articulate what you looked at, what you found, what it means, and what to do about it.
- Self-directed — investigation requests are rarely fully specified. You know how to scope, prioritise, and execute your own work without waiting to be told exactly what to do.
Why Us
The data is irreplaceable. You are working with real sensor streams from operational industrial equipment at enterprise clients. This data does not exist in any public dataset. The patterns you find have never been found before.
The questions are genuinely hard. This is not a business intelligence role. You are not building sales dashboards or cohort reports. You are investigating open-ended problems in complex multivariate time-series data where the answer matters operationally. That is a different level of analytical challenge.
The domain is learnable here. Our team includes domain engineers with deep equipment knowledge, AI-powered investigation tools, and a platform built specifically to make industrial data queryable. You will pick up the industry context faster here than anywhere else — and you will learn it through the data.
The findings have real consequences. When you identify a performance degradation or an anomalous operating condition, that finding goes to an engineer making a decision on an asset worth hundreds of millions of dollars. The stakes make the work matter.
The team is honest. You will work alongside a Data Ops & Analytics Lead (also being hired) who owns the workflow and priorities for the analytics function, domain engineers who will challenge your findings with real-world context, and a Data Science team that values rigour over speed. Feedback is direct.
What We Offer
- Direct access to operational sensor data from 15+ enterprise clients — datasets you cannot get anywhere else
- Close collaboration with domain engineers who carry deep equipment knowledge
- AI-powered tools and a platform built to make industrial data investigation faster
- Real impact: your analyses inform engineering decisions on live equipment
- Hybrid flexibility: 2–3 days on-site in Gurugram, rest remote
- Competitive compensation
Qualifications
- B.Tech / B.E. / B.S. in any engineering, science, mathematics, or quantitative discipline
- 2+ years of hands-on data analysis experience
- Startup or high-growth environment experience preferred
When you apply, show us investigations you have led, not just dashboards you have built.
How to apply for this opportunity
- Step 1: Click On Apply! And Register or Login on our portal.
- Step 2: Complete the Screening Form & Upload updated Resume
- Step 3: Increase your chances to get shortlisted & meet the client for the Interview!
About Uplers:
Our goal is to make hiring reliable, simple, and fast. Our role will be to help all our talents find and apply for relevant contractual onsite opportunities and progress in their career. We will support any grievances or challenges you may face during the engagement.
(Note: There are many more opportunities apart from this on the portal. Depending on the assessments you clear, you can apply for them as well).
So, if you are ready for a new challenge, a great work environment, and an opportunity to take your career to the next level, don't hesitate to apply today. We are waiting for you!
