// kpriyadharshan.dev
Kannan
Priyadharshan
AI Engineer · Agentic Infrastructure
Building agentic AI systems that think, act, and integrate. CS @ NTU 2029. AI Engineer @ Cyber Sierra.
02 · Selected Work
Projects that
mean something
01 · FEATURED
trace-zeroFull-stack optimal execution simulator implementing Almgren-Chriss (2000) against real Binance L1 orderbook data. Compares AC optimal trajectory vs TWAP vs Market Dump across isolated exchange instances — each strategy gets its own SimulatedBook so permanent price impact never contaminates competing lanes. Measures implementation shortfall in bps. Bloomberg Terminal-aesthetic UI.
Python · FastAPI · Next.js · NumPy · WebSocket
View on GitHub →02 · INFRASTRUCTURE
market-replayNanosecond-accurate top-of-book capture and replay for execution backtests. Monotonic clock timestamps — immune to DST and NTP adjustments — as the authoritative timing source. Pluggable handler system, JSONL storage with 10-minute file rotation.
Python · WebSocket · Binance
View on GitHub →03 · QUANT
capm-portfolio-optimizerBeta estimation via linear regression against a market index. Constructs the optimal portfolio using Markowitz Mean-Variance Optimisation. Monte Carlo simulation across 10,000 weight combinations to map the efficient frontier and identify the max Sharpe ratio portfolio.
Python · NumPy · SciPy
View on GitHub →04 · HACKATHON
wayfinderAI career navigation engine. Processes user profiles against career databases via pgvector semantic search, generating personalised 4-week upskilling roadmaps. Top 13 / 100+ teams at NTU Techfest 2026.
Next.js · pgvector · OpenAI · PostgreSQL
View on GitHub →05 · NLP
pollpulse-tnReal-time NLP sentiment pipeline for Tamil Nadu 2026 election forecasting. Aggregates and classifies social signals across sources to surface swing-district indicators.
Python · NLP · Supabase
View on GitHub →03 · Experience
Where I've
shipped
AI Engineer Intern
- Designing and building production agentic AI systems on an async Python stack, integrating LLM orchestration with sandboxed tool execution and streaming chat interfaces
- Building semantic retrieval pipelines with vector embeddings and structured eval frameworks to measure and improve agent output quality over time
- Implementing event-driven backend workflows with human-in-the-loop approval gates and third-party integrations across the full stack
Backend Engineer Intern
- Architected high-throughput data pipelines (AWS SQS, Async Python) processing 10,000+ daily events with zero message loss for downstream Generative AI models
- Productionised internal microservices by enforcing CI/CD and mypy, enabling robust Kubernetes deployment of ML-driven features
- Optimised data retrieval latency to <50ms for Discovery AI Core by migrating to gRPC and QUIC, enabling real-time analytics
Software Engineer Intern
- Built scalable ETL pipelines in Python that onboarded 100k+ SKUs from 30+ brands for training of 3D fit recommendation models
- Deployed a transcription model into production backend, reducing processing time by 70% for stakeholders
- Engineered high-concurrency cloud infrastructure on AWS EKS handling 1,000+ concurrent users with low-latency HTTP/3 ingestion
Backend Engineer
- Engineered a document ingestion pipeline for Skills@CCDS that classifies uploaded PDFs (resume vs LinkedIn export) via OpenAI embedding similarity
- Extracted structured profile data using a schema-constrained LLM prompt with Zod validation and skill normalisation (e.g. Go vs Golang)
- Surfaced results via REST endpoint for human-in-the-loop verification before profile commit; PR merged into production codebase
Head of Technology (Backend)
- Led backend development for orientation games platform serving 1,000+ incoming students using Redis and WebSockets for real-time score tracking
- Conducted code reviews and managed deployment schedules across a team of student developers
Director, Quantitative Finance Academy
- Spearheading curriculum design for 80+ members, structuring advanced lessons on Market Microstructure and ML in Finance
- Collaborating with industry partners to facilitate technical workshops on quantitative finance
04 · Now
What I'm
focused on
Agentic AI platform @ Cyber Sierra — multi-agent system for natural language querying of codebases via Slack, chaining skills (/grill-me, /to-prd, /to-issues) and human-in-the-loop verification. Python, FastAPI, Inngest, Slack Bolt, Anthropic SDK.
Zoom Clone in Rust — WebRTC signalling server + SFU from scratch in Rust/tokio
Cursor for PMs — early validation stage. Interviewing product managers to find the one workflow worth eliminating.
05 · Stack
Tools I
know well
06 · Writing
Thinking
out loud
Building a Market Data Replay Engine with Nanosecond Precision
Why monotonic clocks, JSONL, and the writer/reader thread separation matter in market data infrastructure.
07 · About
Who I
am
I build agentic AI systems — the kind that think, act, and integrate with the tools teams actually use. Right now that means designing production multi-agent pipelines at Cyber Sierra, an AI-powered GRC platform in Singapore, where I work on LLM orchestration, semantic retrieval, eval frameworks, and event-driven human-in-the-loop workflows.
Before that: high-throughput data pipelines at eLife (AWS SQS, gRPC, Kubernetes), ETL infrastructure at Netvirta (AWS EKS, 100k+ SKU onboarding), and a document ingestion pipeline for Skills@CCDS — PDF classification via OpenAI embeddings, schema-constrained LLM extraction, and human-in-the-loop verification before profile commit.
At core, I'm a problem solver. The domain matters less than the depth — which is why I've followed the real leverage from quant infrastructure into agentic AI. I'm also Director of the Quantitative Finance Academy at NTU, where I design curriculum on market microstructure and ML in finance for 80+ members.
Outside of this: varsity cricket for NTU, Hiphop Tamizha, and a long-term plan to own a coffee estate in Kodaikanal.
Reading
Get in Touch
Open to internship opportunities for Summer 2027 in Hong Kong and Singapore.
Particularly interested in agentic AI infrastructure, backend systems, and applied AI research roles. If you're building at the intersection of AI and real-world systems, I'd like to hear about it.