# Priyam Jain > Priyam Jain is an AI product engineer and Software Engineer II at Cisco in Bengaluru. He took an enterprise product from inception to production at Cisco, originated and led the open-source project Agentic Lineage, and built Opbase, which reports more than six active clients. He also helped launch TruthSeek, a voice AI startup that earned revenue before closing. [Portfolio](https://priyamjain.me/) · [Profile JSON](https://priyamjain.me/profile.json) · [GitHub](https://github.com/priyam-jain-2002) · [LinkedIn](https://linkedin.com/in/jain-priyam) ## Areas of work - [AI product engineering](https://priyamjain.me/expertise/ai-product-engineering/): I’m an AI product engineer based in Bengaluru and a Software Engineer II at Cisco. My work spans production software for enterprise clients, AI-assisted troubleshooting, and voice AI for customer research. - [AI agent infrastructure](https://priyamjain.me/expertise/agent-infrastructure/): I originated Agentic Lineage and lead its architecture, roadmap and releases. It packages workflows, skills, policies and other assets so developers can inspect and use the same agent setup across projects and coding tools. - [Workflow automation and operations software](https://priyamjain.me/expertise/workflow-automation/): I build software for operational work that has outgrown spreadsheets and messages. With Opbase, the problem was keeping track of industrial orders as they moved between people, channels and partial deliveries. ## Selected work ### [Project Raptor at Cisco](https://priyamjain.me/work/raptor/) **Enterprise production.** Took a Cisco product from inception to production for major enterprise clients. - Engineering impact: Delivered a production service and took responsibility for how it ran after release. - Business impact: Supported production work for major enterprise clients. - My role and ownership: Owned system design, full-stack engineering, deployment and service reliability through production. - Scope and constraints: Published at a high level to protect client and product confidentiality. ### [Cloudshield](https://priyamjain.me/work/cloudshield/) **Production observability.** Helped bring an AI-assisted troubleshooting platform for distributed infrastructure into production. - Engineering impact: Helped bring diagnostic evidence and AI-assisted investigation into a workflow engineers could use in production. - Business impact: Gave engineers a repeatable way to assemble diagnostic evidence for production incidents. - My role and ownership: Contributed to backend engineering, observability and production delivery for AI-assisted investigations. - Scope and constraints: Published at a high level to protect product and operational details. ### [Opbase](https://priyamjain.me/work/opbase/) **Operational use.** Built an industrial order operations platform with more than six active clients, as reported by its operator. - Engineering impact: Made the order record the place to check status, ownership and partial deliveries. That gave staff a shared record for day-to-day operations, documentation and analysis. - Business impact: Opbase has more than six active clients, as reported by its operator. In one operation using it, the team reported zero missed orders and zero missed compliance steps, with routine coordination down to a minimum. Annual documentation and analytics work fell from several weeks to about two days. - Problem: Orders came in across channels. Staff had to reconcile handoffs, partial deliveries and records across messages and spreadsheets, then reconstruct that history for reporting. - My role and ownership: Designed the workflow and built the operational software linking order intake, catalog, inventory and fulfillment. - Architecture and tradeoffs: NestJS and PostgreSQL support a shared order record and explicit transitions. React gives owners, store managers and order managers different views of the same work. Webhooks and notifications connect channel events and operational handoffs. - Workflow in use: Capture an incoming order, assign it, track status and partial deliveries, and surface open items to the next owner. - Scope and constraints: The active client count is a firsthand report from November 2026. The operating improvements describe one deployment. - [Opbase overview on Pinchit](https://www.pinchit.in/#work) ### [TruthSeek](https://priyamjain.me/work/truthseek/) **Closed startup · historical case study.** Helped launch an AI voice interview product that reached an enterprise pilot and earned revenue before closing. - Engineering impact: Connected study setup, voice interviews and post-interview review into a working customer workflow. - Business impact: TruthSeek served customers, reached an enterprise pilot and generated revenue during its operating period. The startup has since closed. - Problem: Research teams needed to collect customer interviews consistently and review the raw material after each call. - My role and ownership: Helped build the product from concept through an enterprise pilot, working across the application and voice interview workflow. - Architecture and tradeoffs: The Next.js application and NestJS backend connected study setup and review to the Vapi voice integration. The stack used Supabase and OpenAI. Recordings and transcripts gave researchers material to review alongside the analysis. - Workflow in use: Define a research brief, prepare questions and an audience, run interviews, then review recordings, transcripts and analysis. - Scope and constraints: The linked product page documents the historical workflow. - [Historical product walkthrough](https://truthseek.in/) ### [Agentic Lineage](https://priyamjain.me/work/agentic-lineage/) **Open source.** Created Agentic Lineage and lead its development. The open-source Go runtime packages and distributes agent workflows across coding tools. - Engineering impact: Made agent environments inspectable, versionable and reusable across projects, with provider-specific setup behind an adapter boundary. - Business impact: Technical leads discussed how to reuse shared workflows. The repository and releases show the delivered implementation. - Problem: Agent workflows depend on assets that are easy to lose when copied between projects: skills, policies, workflow steps and provider configuration. - My role and ownership: I conceived Lineage and lead its technical direction, roadmap and releases. I designed the package and execution model and documented a review standard that asks contributors to link an issue and explain how their change is tested. I also presented the project and discussed practical reuse with technical leads. - Architecture and tradeoffs: A manifest declares package exports and capabilities. Deterministic archives carry the assets, and the receiver verifies content digests before use. The registry stores versioned artifacts while local provider adapters materialize the workflow for tools such as Claude Code and Codex. - Workflow in use: Package and publish a workflow, resolve a version, inspect its contents, then enable it and run it through a provider adapter. - Scope and constraints: The product exposes declared capabilities for inspection. Those declarations do not enforce runtime isolation. - [Source code](https://github.com/agentic-lineage/lineage) - [Architecture](https://github.com/agentic-lineage/lineage/blob/main/docs/architecture.md) - [Pull request review standard](https://github.com/agentic-lineage/lineage/blob/main/docs/decisions/0004-require-issue-links-and-test-plans-for-prs.md) - [Releases](https://github.com/agentic-lineage/lineage/releases) ### [Pinchit](https://priyamjain.me/work/pinchit/) **Active automation studio.** Built an automation studio that maps operational bottlenecks and ships software around them. - Engineering impact: Made responsibilities and exception paths explicit in the workflows Pinchit builds. Opbase shows how that approach applies to orders, inventory and supplier coordination. - Business impact: Pinchit publishes examples of completed client work, including Quarks Industrials and TourWhisker. - Problem: Repeated checks, copying and handoffs leave teams with avoidable labor and unclear ownership. - My role and ownership: Developed the studio’s delivery approach from workflow discovery to implementation and follow-up measurement. - Architecture and tradeoffs: Start with the order, audit or back-office workflow. Map where data enters, who can approve an action and what happens when a case is unusual. Automate stable processing steps and route exceptions to a person. - Workflow in use: Measure the existing workflow, build the smallest change, put it in the hands of operators, and compare cycle time, rework and cost per case. - Scope and constraints: Individual contract terms and commercially sensitive results are private. - [Studio and project examples](https://www.pinchit.in/) ### [AI for network investigations at Cisco](https://priyamjain.me/work/cisco-network-investigation/) **Escalation team evaluation.** Built tooling to help Cisco escalation engineers investigate network anomalies and possible root causes. - Engineering impact: Delivered a working investigation flow to an escalation team for evaluation. - Business impact: The evaluation targeted the time engineers spend collecting and reviewing diagnostic evidence. - My role and ownership: Delivered the first version for evaluation by an escalation team in about a month. - Scope and constraints: Published at a high level to protect investigation and account details. ### [Multi-Agent Capital Pipeline](https://priyamjain.me/work/market-research-engine/) **Implemented · paper trading by default.** Built an Indian equities research engine with eight parallel agents, including an agent for geopolitical signals. - Engineering impact: Connected research, risk decisions, simulated execution and performance review in one workflow. - Business impact: The system supports strategy evaluation with paper trading as the default. - Problem: Fundamentals, liquidity, governance and geopolitical events can point in different directions. A useful workflow has to preserve the evidence behind a recommendation and constrain portfolio exposure. - My role and ownership: Built the research pipeline that combines independent signals, applies portfolio risk rules and exposes position and performance data in a dashboard. - Architecture and tradeoffs: Python runs eight analyses concurrently and collects a common result with confidence and risk flags. An aggregation stage checks agreement, then a market-regime stage adjusts sizing. Portfolio constraints sit between signal generation and the broker adapter. FastAPI serves the dashboard. - Workflow in use: Scan the configured universe, review signals and risk flags, simulate orders, and review open positions and performance on the dashboard. - Scope and constraints: This is the Multi-Agent Capital Pipeline, a separate project from JFT. Realized performance after costs has not been verified for publication. ### [KubeVision at Cisco](https://priyamjain.me/work/kubevision/) **Internship project.** Built Kubernetes monitoring and pod visibility tooling during the Cisco internship. - My role and ownership: Used Python, Docker, and React and explored anomaly detection for spikes in cluster telemetry. Presented the work to senior directors. ### [Rust Proof-of-Stake Escrow](https://priyamjain.me/work/rust-escrow/) **Testnet prototype.** Built a Rust blockchain prototype for programmable escrow and deployed it to a test network. - My role and ownership: The project explored how software can encode escrow conditions and release funds when those conditions are met. ### [JFT — Juggad Frequency Trading](https://priyamjain.me/work/jft/) **Research and system design.** Designed a phased research system for studying short-horizon price signals in Indian markets. - My role and ownership: Scoped a 30-second research cadence, a Rust implementation, and a broker interface designed to support adapters such as Zerodha. ## Project catalog [Indexable project pages](https://priyamjain.me/catalog/) - [AI Collaboration & Task Summarization Agents](https://priyamjain.me/catalog/HSZU5rUAvPHLlUSR1ZK1/) - [AI-Based Kubernetes Anomaly Detection System](https://priyamjain.me/catalog/zGVucLVXXLaMrSCOn6lA/) - [Blockchain Escrow — Trustless Payment & Commitment Layer](https://priyamjain.me/catalog/IaRdQp3kxFTmTRw2GmMj/) - [Pinchit — Proof-of-Work Project Marketplace](https://priyamjain.me/catalog/o6JS8N0ayysiJcn0xb0Q/) - [Reputation Engine — Proof-of-Work Scoring System](https://priyamjain.me/catalog/fpQGFZg4T0XEDU0mPJic/) - [Self Improving agents](https://priyamjain.me/catalog/SgTu65GD8uF9EMx56qVa/) [Browse all published project records](https://priyamjain.me/project-catalog.txt) for project descriptions, technical details, and links. The catalog is generated from the portfolio's Firestore data. ## Writing [Writing archive](https://priyamjain.me/blog/) - [Building AI Applications with the Vercel AI SDK](https://priyamjain.me/blog/BzAqfUV1mKZmNn88L7I1/): A deep dive into creating intelligent applications using the latest AI tools and frameworks - [Smart Contract Security: Best Practices for DeFi](https://priyamjain.me/blog/q4jjEH7aMFSSAXpG0oAJ/): Essential security patterns and common vulnerabilities to avoid when building DeFi protocols - [React Server Components: The Future of Web Development](https://priyamjain.me/blog/kVPXxcotDt1Ulziw0JER/): Understanding the paradigm shift and how to leverage server components for better performance ## Experience [Engineering experience](https://priyamjain.me/experience/) - Software Engineer II, Cisco (2024-08 to present). Took Project Raptor from inception to production for major enterprise clients, owning system design, full-stack delivery, deployment and service reliability. Built AI-assisted network investigation tooling for evaluation by a Cisco escalation team. - Technical Intern, Cisco (2024-01 to 2024-06). Built KubeVision, a Kubernetes monitoring and visibility project, and explored anomaly detection for Kubernetes telemetry. Presented the work to senior directors. - Best Contributor recognition at Cisco - Cisco hackathon wins - Bachelor of Technology in Computer Science and Engineering (Artificial Intelligence and Machine Learning), VIT Bhopal University (2020–2024) ## Skills - languages: Python, TypeScript, Go, Rust, JavaScript, C++ - AI and data: LLM applications, LangChain, MCP, Anomaly detection, Natural-language data interfaces - backend: NestJS, FastAPI, PostgreSQL, Redis - frontend: React, Next.js - infrastructure: Docker, Kubernetes, Jenkins, Ansible, Prometheus, Grafana