I build machine-learning systems and run them on real infrastructure. The thread through my work is restraint: a rules layer instead of a rebuilt scoring model, a human approving every automated dispatch, fairness reported as a full range instead of one flattering number. I would rather ship the version that holds up than the version that only demos well.
A copilot for 911 dispatch: it transcribes incoming calls, clusters duplicate reports into single incidents, and recommends which units to send.
▲ Team radio · the call
The system only recommends — a human dispatcher approves every dispatch. The optimizer re-plans the moment a higher-priority call lands, instead of treating allocation as a one-shot decision.
45 units and 3 hospitals simulated with live capacity tracking; 18 tests; built in two days.
45 units
3 hospitals
2-day build
18 tests
A human approves every dispatch
FastAPIFastAPI A Python framework for building web APIs.
SQLAlchemySQLAlchemy A Python toolkit for working with databases.
ReactReact The most widely used library for building web interfaces.
LeafletLeaflet A JavaScript library for interactive maps.
WebSocketsWebSockets A live, two-way connection between a browser and a server.
Docker ComposeDocker Compose Runs several Docker containers together as one app.
LLMsLLMs Large language models: the kind of AI behind chat assistants.
P2Team lead · graduate ML · 2026
Fairness stress-testing framework
A framework that tests whether a published fairness fix actually holds up when you change the random seedrandom seed The starting number for a model's randomness; changing it can quietly change the results., the decision thresholddecision threshold The cut-off score above which a model says yes., or the subgroup you look at.
▲ Team radio · the call
Reported the full range of results across random seeds and thresholds instead of the usual single flattering number — and showed that how fragile a fix is depends on the model.
Threshold choice alone swung the fairness gapfairness gap A fairness measure: the difference between groups in how often a model gives the positive outcome. 0 means every group gets it at the same rate. 7× (0.04 → 0.28), as much as switching the debiasing methoddebiasing method A technique applied to a model to reduce unfair differences between groups..
7×Fairness gap · threshold alone
0.040.280.000.100.200.30
scikit-learnscikit-learn The standard Python library for classic machine learning.
XGBoostXGBoost A popular library for gradient-boosted decision trees, strong on tabular data.
FairlearnFairlearn An open-source toolkit for measuring and improving fairness in machine-learning models.
pandaspandas Python's main library for working with tables of data.
NumPyNumPy Python's core library for fast numerical arrays.
P3Applied cryptography · 2026
Secure login: TLS 1.3 + OPAQUE, from scratch
A TLS 1.3TLS 1.3 The protocol behind the padlock in your browser: it sets up an encrypted connection to a website. 1.3 is the current version.–style secure connection built from scratch in Python, with OPAQUEOPAQUE A modern password-login protocol: the server can check your password without ever seeing it or storing it. password login in place of the usual Diffie-HellmanDiffie-Hellman The classic way for two computers to agree on a secret key over an open network. key exchange: the server never sees, stores or can pre-compute against your password.
▲ Team radio · the call
Ran the OPRFOPRF Oblivious pseudorandom function: the server helps turn the password into a key without learning the password. on plain Ed25519Ed25519 A widely used elliptic curve, behind many digital signatures and SSH keys. curve operations instead of the textbook Ristretto255Ristretto255 A safer way of using the same curve that removes a known pitfall (the cofactor); the textbook choice for OPAQUE. — safe here because every OPRF output is hashed, so no raw curve point is ever exposed.
A wrong password fails on your own device and leaks nothing to the server. 12/12 tests pass.
ClientPassword
✕
Server
never sees it
never stores it
never pre-computes against it
12/12 tests pass
PythonPython The default language for machine learning, data and back-end work.
Ed25519Ed25519 A widely used elliptic curve, behind many digital signatures and SSH keys.
X25519X25519 Diffie-Hellman key agreement on Curve25519, the standard in modern TLS.
AES-GCMAES-GCM AES in a mode that both encrypts data and detects any tampering with it.
HKDFHKDF A standard way to turn one shared secret into several separate keys.
X.509X.509 The certificate format websites use to prove who they are.
P4Software Engineer Intern · 2026
Customer-health triage engine
A triage layer on top of an existing customer-health score that tells success managers which accounts need attention, why, and what to do next.
▲ Team radio · the call
Built it as deterministic rules, not ML — the logic didn't need a model, and a model would have been harder to trust and to explain to the people acting on it.
Surfaced quietly at-risk accounts the existing score never flagged.
Customer-health score
Deterministic rules
Which accounts
Why
What next
Rules, not ML: easier to trust and explain
PythonPython The default language for machine learning, data and back-end work.
pandaspandas Python's main library for working with tables of data.
S3
Experience
Strategy chart each stint, by year
2023202420252026JobCadetFounder2023–2491mobilesHardware Benchmarking, Central Testing Team2023–24Times InternetAI / ML Engineer2024KLJ GroupAI Engineer2025ParentSquareSoftware Engineer Intern2026ASU Enterprise TechnologyStudent Engineer, Engineering Technology2026 → now
DevOpsDevOps Automating how software is built, tested, shipped and run. and infrastructure work in the university's central technology organization, inside its GitHub org and AWS environment.
2026
Software Engineer Intern · ParentSquare
Built the customer-health triage engine above, plus CRUDCRUD Create, read, update, delete: the basic operations behind most app features. feature work on the Digital Business Review product in React and TypeScript on a pull-request CI/CDCI/CD Automated testing and deployment on every code change. workflow.
2025
AI Engineer · KLJ Group
Built and ran LLMLLM Large language models: the kind of AI behind chat assistants.inferenceinference Running a trained model to get its answers, as opposed to training it. pipelines for three products — billing automation, an email-thread reader, and voice + vision search — with Python and Go on the serving path and local and cloud models served through vLLMvLLM An open-source engine for serving large language models quickly and cheaply..
2024
AI / ML Engineer · Times Internet
Shipped ML inferenceinference Running a trained model to get its answers, as opposed to training it. APIs and content-classification pipelines; KafkaKafka A system for streaming events between services reliably. for async events, RedisRedis A very fast in-memory data store, often used as a cache. for caching, Docker and CI/CDCI/CD Automated testing and deployment on every code change., deployed across AWS and GCP.
2023 – 2024
Hardware Benchmarking, Central Testing Team · 91mobiles
Benchmarked and analyzed phones across CPU, GPU, memory and thermals, and trained new interns on the process.
2023 – 2024
Founder · JobCadet
AI recruitment startup; raised ₹6 lakh in NASSCOMNASSCOM India's national IT-industry association. seed funding.
S4
Research
2023Metal–organic frameworks
2024Human activity recognition
2024Speech emotion recognition
Detection of Open Metal Sites in Metal–Organic Frameworks Using Machine Learning
Vision-Based Human Activity Recognition Using CNNCNN Convolutional neural network: a model type built for images and signals. and LSTMLSTM A neural network with memory, suited to sequences such as video frames or speech. Architecture
Ranked #1 in India — led backend work turning NASA datasets into visualizations.
2
1
3
P1 · India
2023
BVEST Hackathon
Led a 100-student organizing team; 350+ participants and ₹3 lakh in sponsorship raised.
100 organizing team
350+ participants+
₹3 lakh sponsorship · one mark = ₹1 lakh₹₹₹
2022
ISTE Club — General Secretary
Led 150+ students across events and technical initiatives.
150+ students led+
S6
Skills
Proof map every skill, wired to where it was used · hover a skill or a place
Hover a skill to see what it is and exactly where it was used. Hover a place to see everything it proves; a bigger dot proves more.
PythonThe default language for machine learning, data and back-end work.Proven at: KLJ pipelines · fairness framework · triage engine
GoGoogle's language for fast, simple network services.Proven at: KLJ serving path
CA low-level language that runs close to the hardware, used here for raw speed.Proven at: AES cryptanalysis
React · TypeScriptThe most widely used library for web interfaces, written in typed JavaScript.Proven at: ParentSquare DBR · Atlas
FastAPIA Python framework for building web APIs.Proven at: Atlas
vLLM / LLM servingRunning large language models in production; vLLM is an open-source engine that serves them fast.Proven at: KLJ inference products
PyTorch · scikit-learn · XGBoostPython's core machine-learning toolkits: deep learning, classic ML and gradient-boosted trees.Proven at: research · fairness framework
DockerPackages an app with everything it needs, so it runs the same anywhere.Proven at: Times Internet · Atlas
Kafka · RedisKafka streams events between services; Redis is an in-memory store, often used as a cache.Proven at: Times Internet pipelines
AWS · GCPAmazon's and Google's cloud platforms.Proven at: Times Internet deployments
Applied cryptographyBuilding real security (encryption, key exchange, secure login) and breaking weakened versions of it.Proven at: TLS 1.3 + OPAQUE · AES CTF
D3.jsA JavaScript library for interactive data visualisation.Proven at: interactive data-viz coursework
S7
About
Team sheet off the track
Home
New Delhi → Tempe
Favourite player
Messi · craft over talent
In three words
Loyal · tenacious · argumentative
Off the keyboard
Barcelona · Vettel-era F1 · anime and manga
I grew up in New Delhi and moved around a lot, but went to the same school the whole way through — one fixed point while everything else changed. Now I'm in Tempe for an MS in Computer Science at Arizona State.
My favourite player is Messi, and the reason is the honest one: in raw ability he isn't on another planet, he's just outworked everyone for twenty years. Craft over talent. That's the standard I hold my own work to.
Three words my friends use for me: loyal, tenacious, argumentative. I ship fast, I push back when something's too safe, and I'd rather hand you a working prototype than a document describing one.
Away from the keyboard: Barcelona, Vettel-era F1, and more anime and manga than I'll admit on a professional site.
The protocol behind the padlock in your browser: it sets up an encrypted connection to a website. 1.3 is the current version.A modern password-login protocol: the server can check your password without ever seeing it or storing it.The classic way for two computers to agree on a secret key over an open network.Oblivious pseudorandom function: the server helps turn the password into a key without learning the password.A widely used elliptic curve, behind many digital signatures and SSH keys.A safer way of using the same curve that removes a known pitfall (the cofactor); the textbook choice for OPAQUE.Diffie-Hellman key agreement on Curve25519, the standard in modern TLS.AES in a mode that both encrypts data and detects any tampering with it.A standard way to turn one shared secret into several separate keys.The certificate format websites use to prove who they are.The Advanced Encryption Standard, the cipher used almost everywhere. A full 128-bit key is far beyond brute force.AES-128 scrambles data in 10 rounds; a reduced-round version runs fewer, which makes it breakable. The usual setup for a security challenge.Capture the flag: a security competition where you break deliberately weakened systems.2 to the power of 24: 16,777,216 possibilities.A fairness measure: the difference between groups in how often a model gives the positive outcome. 0 means every group gets it at the same rate.The cut-off score above which a model says yes.A technique applied to a model to reduce unfair differences between groups.The starting number for a model's randomness; changing it can quietly change the results.Running a trained model to get its answers, as opposed to training it.Large language models: the kind of AI behind chat assistants.An open-source engine for serving large language models quickly and cheaply.Convolutional neural network: a model type built for images and signals.A neural network with memory, suited to sequences such as video frames or speech.Porous crystalline materials built from metal ions linked by organic molecules.An open-source toolkit for measuring and improving fairness in machine-learning models.A popular library for gradient-boosted decision trees, strong on tabular data.A system for streaming events between services reliably.A very fast in-memory data store, often used as a cache.Automated testing and deployment on every code change.Automating how software is built, tested, shipped and run.Create, read, update, delete: the basic operations behind most app features.India's national IT-industry association.A Python framework for building web APIs.A live, two-way connection between a browser and a server.A JavaScript library for interactive maps.A Python toolkit for working with databases.A JavaScript library for interactive data visualisation.The default language for machine learning, data and back-end work.Google's language for fast, simple network services.The most widely used library for building web interfaces.JavaScript with types, which catches mistakes before the code runs.Runs several Docker containers together as one app.Packages software with everything it needs, so it runs the same anywhere.Python's main library for working with tables of data.Python's core library for fast numerical arrays.The standard Python library for classic machine learning.A leading deep-learning framework.The default language for machine learning, data and back-end work. Proven at: KLJ pipelines · fairness framework · triage engine.Google's language for fast, simple network services. Proven at: KLJ serving path.A low-level language that runs close to the hardware, used here for raw speed. Proven at: AES cryptanalysis.The most widely used library for web interfaces, written in typed JavaScript. Proven at: ParentSquare DBR · Atlas.A Python framework for building web APIs. Proven at: Atlas.Running large language models in production; vLLM is an open-source engine that serves them fast. Proven at: KLJ inference products.Python's core machine-learning toolkits: deep learning, classic ML and gradient-boosted trees. Proven at: research · fairness framework.Packages an app with everything it needs, so it runs the same anywhere. Proven at: Times Internet · Atlas.Kafka streams events between services; Redis is an in-memory store, often used as a cache. Proven at: Times Internet pipelines.Amazon's and Google's cloud platforms. Proven at: Times Internet deployments.Building real security (encryption, key exchange, secure login) and breaking weakened versions of it. Proven at: TLS 1.3 + OPAQUE · AES CTF.A JavaScript library for interactive data visualisation. Proven at: interactive data-viz coursework.