Timur Golovinov - AI & Software Engineer and engineering lead

Timur Golovinov

AI & Software Engineer · Engineering lead

I turn enterprise problems into elegant AI products, and lead the engineers who ship them. Throughout my career I've built products that combine AI, high-performance browser graphics, geospatial and wearable data, and cloud infrastructure, from deep technical full-stack work to leading teams on business-critical launches.

At Jasper (AI for enterprise marketing, used by 20%+ of the Fortune 500), I was on the team the company exclusively trusted with its p0s. I owned Studio, Jasper's no-code agent builder, end to end, and built core Task Agents infrastructure: context engineering, enterprise governance, evals and long-running async workflows. Most recently I led the AI research team as acting engineering manager, working out how AI assistants find, describe and recommend brands.

My favourite tools are TypeScript, Node, React and Python on GCP and AWS, with LLM APIs from OpenAI, Anthropic and Google. I pick whatever fits the problem, and I work with coding agents daily, holding their output to the same review bar as hand-written code.

Outside of work, I enjoy kitesurfing and exploring the intersection of human behaviour and technology. I study neuroscience and psychology, which informs how I design intuitive, user-centred products.

Experience

Jasper logo

Senior Software Engineer & Acting Engineering Manager

Jasper

Australia

Applied AI company serving 20%+ of the Fortune 500. I worked on its three most business-critical areas: the agent platform, the Studio agent builder, and its 2026 AI research push.

LLM APIs (OpenAI, Anthropic, Google)AgentsContext EngineeringEvalsMCPRAGTypeScriptPythonNestJSReact / Next.jsGraphQLPostgreSQLPrismaTemporalGCPDockerDatadogPlaywrightCoding agents (Claude Code, Cursor)
AI research · acting engineering managerAug – Nov 2026

Led 4 senior and staff engineers on the company's p0: helping enterprises win visibility in AI answers, a new field with no playbook.

  • Co-founded a self-improving research loop: agents draft and run pre-registered experiments daily, a second AI reviews them sceptically, a human approves, and each finding seeds the next hypotheses. 39 experiments and 128 hypotheses decided in ~4 weeks.
  • Changed how product and engineering decide: every recommendation now has to be backed by evidence. Findings also became GTM material and playbooks for forward-deployed engineers.
  • In one quarter the team shipped 3 projects and ~14 customer launches, 4 of which I built.
Studio, Jasper's agent builder2026

Sole engineer and lead, effectively a one-person EPD team (product, design and engineering) for a no-code agent builder running ~11k enterprise agents.

  • Ran calls with top enterprise accounts and shipped 17 enhancements they asked for in six months, including agent version control for a Fortune 50 customer's governance rollout.
  • Built context engineering, agent testing and verification, and an "Auto" mode that keeps every agent on the current best model across 8 frontier models.
  • Instrumented it end to end with observability, quality and usage metrics, so the roadmap came from data and customer calls.
  • After I moved to lead the AI research team, Studio became the company's p1 (behind GEO as p0), and a full team took over what I had built and run alone.
Agents, the core platform primitive2023 – 2026

Jasper's main high-throughput primitive, behind the document editor, Canvas, Studio, the Public API and integrations, and the base Solutions and forward-deployed engineers built customer workflows on.

  • Core agent primitives: the harness, context engineering, governance, evals, and long-running async workflows.
  • Agent UI across surfaces: Canvas, chat, dashboards and generative UI.
  • Cut agent workflow failures from ~20% to under 3% in ~6 weeks: observability first, then root causes.
  • Architecture for scale: designed the Tasks platform that became the agent builder's foundation, led the Canvas Public API to production with zero breaking changes, and migrated ~6k production prompt versions with zero incidents.
Engineering foundations

Deep engineering work that kept the platform fast and the team moving.

  • Improved performance by orders of magnitude: O(n^2) to O(n), GraphQL n+1 fixes, Prisma-generated SQL cut from 250+ queries to 10; found and fixed memory leaks.
  • Designed and built, to PR stage, a pub/sub layer connecting real-time collaborative documents (Yjs CRDT) with long-running agent runs and the Public API, with very little code.
  • Changed company-wide code review from two required reviewers to one, as a reversible, compliance-checked experiment.
  • AI-native: 63% of my 2026 commits were co-authored with an AI coding agent, reviewed to the same bar as my own code.
Technical Challenges
  • Non-deterministic AI, and measuring it reliably
  • A new field with no playbook: winning visibility in AI answers (GEO)
  • Reliability of long-running agent workflows at enterprise scale
  • Balancing velocity and quality in a fast-moving AI market
SpaceDraft logo

Software Engineer

SpaceDraft

Australia

Led full-stack development of core features including Images, Video, Audio, Maps with data layers. These features directly landed >50% of high-value enterprise, education and government customers, additionally securing crucial funding from the Australian government and private investors.

ReactReduxTypeScriptJavaScriptNodeRESTGraphQLNoSQLAWS (15+ services)CI/CDWeb APIsArcGISHTMLCSSCypressGit
Overview▼

I led full-stack development of core features including Images, Video, Audio, Maps with data layers. These features directly landed >50% of high-value enterprise, education and government customers, additionally securing crucial funding from the Australian government and private investors.

I designed and developed multiple BE and FE core features of the product, including geospatial data visualisation (as per regulated gov grant requirement), video streaming / video-on-demand using AWS services, which played a critical role in the app's usability and bandwidth consumption, and multitrack audio recording, playback and basic editing capabilities, ensuring accessibility, interoperability, scalability and maintainability.

I designed and developed a paywall system for a SaaS business, integrating Stripe deeply into a web app and website to automate complex transactions (including refunds, subscription management, and late payments) and streamline financial management.

Key Responsibilities
  • Participate in full product development lifecycle: concept, design, build, deploy, test and release, measure and iterate
  • Cloud architecture
  • Stakeholder liaising and communication
  • Feature development and maintenance
  • Enforcing high standards of code quality & best practices
  • Writing tests
  • Performance optimization across the stack
  • Researching solutions to technical challenges
  • Communicating with stakeholders
  • Participating in project management & planning
  • Mentoring and training junior engineers
Technical Challenges
  • Device and browser interoperability for high-performance application with multiple canvas layers (Animation, Geospacial 3rd Party API, Asset Layers) and web APIs (such as WebRTC)
  • Developing a complex permission system for video-on-demand feature with multiple roles (e.g. admin, editor, viewer) and multiple levels of access (e.g. read, write, delete)
  • Implementing a paywall with Stripe with multiple payment options (e.g one-off, sunscriptions, etc) and classes of users (e.g. free, premium, enterprise)
  • Maintaining and refactoring existing codebase with limiting black-box solutions (such as early versions of AWS amplify)
  • Implementing new features with limited time and resources, strong balance between quality and speed
  • Working in ambiguous environments with limited requirements and frequent changes
  • Implementing resource-intensive features with limited resources (e.g. CPU, GPU, RAM, etc)
  • Refactoring system from being CPU bound (too many things causing lags) to be real-time-bound (e.g. reduce fps when system overloads)
  • Pioneering unprecedented features (multimedia playback and view, geospatial map, animating canvas (time + space), etc) with limited documentation and examples
Key Achievements
  • Developed robust video-on-demand capabilities (serveless system for video upload, transcoding for adaptive bitrate + mono format, artifact generation, manifest editing for permissions) with a sophisticated permissions framework
  • Implemented a geospatial map solution for a government grant that became a core feature of the app (included performance optimisation of ArcGis solutions to enable support of older devices, bundle optimisation, and multiple features such as FeatureLayers, Places, Routes, geocoding, 3rd party data integration, basemap changes, screenshoting and many others)
  • Developed multitrack audio playback and editing using Web Audio API (WebRTC, RecordRTC library), and Audio Context
  • Made significant contributions to the design and development of the app's UI/UX
  • Implemented a paywall with Stripe
  • Developed numerous features based on user feedback
  • Architected and implemented numerous fullstack features
  • Improved the app's performance and scalability by making significant refactoring and improvements to the codebase
  • Designed and implemented multi-track audio playback and editing
  • Optimised multiple processes of app's lifecycle such as build (optimise bundle size), deploy (decrease manual operations needed for AWS deployment), and release
GeoMoby logo

Software Engineer

GeoMoby

Australia

Led the front-end development of a proprietary geofencing and life tracking platform, architected and developed a secure, optimised, and scalable react frontend (PWA) that was integrated with backend and native mobile apps.

ReactTypeScriptMaterial UIOpenLayersREST APISASSOOPFPGitJirafp-tsPWA
Overview▼

I worked on web-based software that combines geospatial intelligence with biological signals, acquired through custom wearables + phones.

This combination of hardware and software allowed us to bring new insights and perspectives for our commercial clients, including innovation branches of high-profile companies.

In addition, our users were people working in high-risk jobs such as emergency response workers and mining workers, whose biological states and physical locations are crucial in risky situations.

Finally, we also created mobile app + webapp that tracks people driving and lifestyle habits with the goal of improving them by providing direct financial incentives. This project was a collaboration with RAC, and you can learn more about it in this article https://geomoby.com/geomoby-and-rac-team-up-to-improve-lives-and-provide-a-positive-impact/

Tech: React frontend + typescript, openlayers.

One of the highlights of my time at GeoMoby was the successful development and deployment of a set of features for a web app that we presented at a major mining emergency response competition. Under a high-pressure environment, I worked with the design and product team to quickly and seamlessly develop a React and AWS solution that was robust enough to impress the end-users and secure two enterprise client contracts. This was a crucial moment for the company's reputation and financial success, and the feedback we received at the event was incorporated into future iterations of our core product.

Key Responsibilities
  • Participate in full app lifecycle: concept, design, build, deploy, test and release
  • Development of new user-facing elements and integration with server-side logic
  • Designing, developing, and maintaining efficient, reusable, and reliable library of React components
  • Optimisation of the application for maximum speed and scalability
  • Implementation of security and data protection
  • Help to improve code quality through writing unit tests, automation and performing code reviews
  • Participate in brainstorming sessions and contribute innovative and original ideas to GeoMoby's technology, algorithms and product
  • Work with the product and design team to understand end-user requirements and use cases, then translate that into a pragmatic and effective technical solution
  • Write and maintain documentation for all features in development
Technical Challenges
  • Mobile battery-intensive features (24/7 live tracking via wearables and phones) to power behavior-based AI features
  • Develop multiple applications with asynchronous team members and multiple stakeholders
  • Follow a strict development process with a high level of quality assurance as per client's design, security and data-management guidelines
  • Work in ambiguous environments and use creative problem-solving skills to identify complex problems and develop innovative solutions that are relevant to the end-user and their domain
  • Refactor legacy code, then integrate it and develop new high quality code in a fast-paced environment
Key Achievements
  • Developed a set of features for a web app that we presented at a major mining emergency response competition
  • Contributed creative ideas to GeoMoby's technology, algorithms and product, implemented them and presented them to the team
  • Successfully architected and developed a web application's frontend with React and Typescript, which code was reused for mobile application as well (via webview)
  • Independently built a set of reusable React components that were used in multiple projects
Various logo

Engineer & Entrepreneur

Various

Australia

Built multiple successful websites and web apps. Led small teams of engineers and designers. Launched an innovative vending business that broke even in the first three months post-launch and was eventually sold.

JavaScriptTypeScriptReactNodeReact NativeMaterial UIAmplifyRESTPythonGitAWSUI/UXConsulting
Overview▼

As an entrepreneur with a passion for innovation in the fields of human health and wellbeing, I have always been fascinated by the intersection of technology and biology. My interests lie particularly in the areas of brain-computer interfaces (BCI), biodevices, neuroscience, and psychology. I believe that these cutting-edge technologies have the potential to revolutionize the way we understand and improve human health. Over the past three years, I have dedicated myself to a number of personal projects related to these areas, exploring the latest developments and opportunities in the field. My background in business has also given me the experience and skills needed to bring these projects to life, including managing a small vending machine business in Australia which I successfully sold. As I continue to pursue my entrepreneurial journey, I am driven by a desire to make a positive impact on the world and to create innovative solutions that improve people's lives. I am always seeking new challenges and opportunities to learn and grow, and I am excited about what the future holds for the fields of human health and wellbeing.

In addition to my passion for innovation in the field of human health and wellbeing, I have also built a successful career serving private clients in the tech industry. Throughout my career, I have worked with a wide range of clients on projects ranging from web-app development to infrastructure migration to the cloud, and general consulting. My strong technical skills and deep understanding of the latest technologies have made me a sought-after advisor for businesses looking to stay ahead of the curve. I have a proven track record of delivering high-quality work and exceeding my clients' expectations, making me a valuable asset to any team or organization. As a forward-thinking individual, I am always seeking new opportunities to expand my skillset and make a difference in the world. Whether I am working on a personal project in the field of human health and wellbeing or consulting with a private client on a tech-related project, I am driven by a passion for innovation and a desire to make a positive impact on the world.

Key Responsibilities
  • Project design and architecture
  • Team leadership
  • Business development
  • Client management
  • Technical consulting
  • Full-stack development
Technical Challenges
  • Diverse project requirements
  • Team management
  • Business operations
  • Technical consulting
Key Achievements
  • Built nine successful websites and web apps
  • Led teams of three and four people of engineers and designers for four successful projects
  • Launched an innovative vending business that broke even in the first three months post-launch and was eventually sold
  • Worked with various clients across marketing, law, accounting, manufacturing, startups, and SMB sectors
  • Successfully delivered projects on time and within budget

Side projects

Generative image and video platform

2026 · personal project

End-to-end image and video generation and editing on self-hosted open-weight models.

  • Built generation and editing pipelines in ComfyUI, with LoRAs and custom nodes (including Apple Silicon fixes).
  • MLOps on serverless GPUs: cold starts, worker scaling, GPU size vs model size, and workflow optimisation (samplers, steps, async runs).
  • Evaluated models, samplers and LoRAs for quality, speed and cost.
ComfyUIPyTorchServerless GPUsPythonTypeScript

Latest Experiences and Achievements Across the Stack

AI (agents, LLMs, evals)

  • Agent infrastructure in production: context engineering, tool use (MCP), long-running async workflows (Temporal), governance and version control
  • Evals and experiment design: LLM-as-judge, gold sets, pre-registered experiments with confidence intervals
  • Multi-provider LLM systems across OpenAI, Anthropic, Google and Perplexity: batch APIs, prompt caching, model upgrades without migrations
  • Multimodal RAG and knowledge-base retrieval for agents
  • AI-native engineering: 63% of my 2026 commits were co-authored with Claude Code; I run parallel coding agents with the same review bar as hand-written code

Frontend (main tool: React)

  • Architected and built over 10 production React apps
  • Solid experience developing AI-native web applications (e.g streaming, reasoning / CoT, chat systems, asynchronous execution, etc...)
  • Built high-performance apps (e.g that use canvas or expensive geospatial renders)
  • Built diverse types of web apps e.g SPA, PWA, SSG, SSR with NextJS, etc
  • Built apps with complex state management (e.g react-context used in tandem with redux, zustand, support for GQL subscriptions etc...)
  • Solid understanding of react engine, how to optimize it and design well with modern functional patterns
  • Simplified architecture by developing design systems
  • Interoperable development - e.g apps that work on all browsers, devices, older phones, etc
  • Mobile development with react native
  • Good experience with major modern libraries and frameworks (e.g tailwind, material ui, NextJS, etc)
  • Good experience with modern testing frameworks (e.g Jest, Playwright, Cypress, etc)
  • Strong fundamentals (HTTP, CSS, semantic HTML, web security (auth, cors, etc), browser performance optimisation, accessibility and interoperability (i.e mobile-first UI and backwards compatibility)
  • Solid experience setting up FE projects e.g tuning build tools, monorepos and federated modules, lint rules, CI/CD (e.g in aws amplify, git, vercel), observability (e.g datadog, logrocket, etc)
  • Scaled multiple analytics systems as product matures, good understanding how to filter signal from noise in those systems (especially relevant in emerging non-deterministic AI systems)
  • Good taste in UI/UX, good understanding of design principles, human psychology and how to apply them in practice

Backend (main tool: Node.js)

  • Built over 20 Node applications, from simple CLI tools to complex high-load backend services
  • Built Public API for LLM agents
  • Built internal many types of APIs, e.g REST, GraphQL (with resolver best practices), WebSockets, etc
  • Built complex async systems with Node.js
  • Solid experience with modern backend libraries, frameworks and 3rd party tools(e.g NestJS, Prisma, PostgreSQL, Redis, HocusPocus and Yjs, etc)
  • Extensive experience with major AI APIs (e.g OpenAI, Anthropic, Google, Cohere, etc)
  • Deployment (e.g git, AWS, GCP, Docker, tilt, argo etc)
  • Testing
  • Observability and performance monitoring (datadog, internal audit logs)
  • SLOs / SLAs compliance (e.g 99.99% uptime, 99.9% latency, etc)
  • Performance optimisation (e.g from O(n^2) to O(n) optimising ORM queries, etc)
  • Good understanding of system design fundamentals (e.g how to build problem-solution-fit but still scalable systems, trade-off analysis, etc)
  • Collaboration, innovation and prototyping via RFC, one-pagers and PoCs
  • Ran many critical database migrations where performance matters
  • AWS and GCP experience
  • Developed complex fullstack type systems - consistent types from DB down to FE
  • Fluent in writing the most strongly typed code with discriminated unions and typeguards
  • Good understanding how to build scalable type systems using typescript

Other

  • Infra: AWS, GCP, Docker, tilt, argo, git
  • AI: agents (ReAct, MCP), multimodal RAG, evals, LLM APIs (OpenAI, Anthropic, Google, Cohere), LangChain, coding agents (Claude Code, Cursor)
  • DB: PostgreSQL, NoSQL (Mongo, DynamoDB), Vector (ChromaDB)
  • Testing: Jest, Playwright, Cypress, Datadog Synthetic tests
  • Observability, Analytics & User Research: Google Analytics, Amplitude, LogRocket, Datadog

Certificates

Google Cloud - Digital Leader by Google

Google Cloud - Digital Leader

Issuer: Google

Year: 2024

See more certificates▼
NCAN Certificate neurotechnology by National Centre for Advanced Neurotechnology

NCAN Certificate neurotechnology

National Centre for Advanced Neurotechnology

2021

Gtec Certificate Neurotechnology by Gtec

Gtec Certificate Neurotechnology

Gtec

2021

Perth biodesign Certificate by Perth Biodesign

Perth biodesign Certificate

Perth Biodesign

2021

Brain Computer Interface Certificate by UDEMY

Brain Computer Interface Certificate

UDEMY

2020

Neural signal processing and analysis Certificate by UDEMY

Neural signal processing and analysis Certificate

UDEMY

2021

Curtin University Masters Degree of Internet Communications by Curtin University

Curtin University Masters Degree of Internet Communications

Curtin University

2021

Diploma in Software Development by TAFE

Diploma in Software Development

TAFE

2019

Certificate IV in Programming by TAFE

Certificate IV in Programming

TAFE

2018

Nest JS Certificate by NestJS

Nest JS Certificate

NestJS

2023

SQL Certificate by UDEMY

SQL Certificate

UDEMY

2024

Relational database design Certificate by UDEMY

Relational database design Certificate

UDEMY

2024

Senior web developer certificate by UDEMY

Senior web developer certificate

UDEMY

2020

Design rules: Principles + Practices for Great UI Design by UDEMY

Design rules: Principles + Practices for Great UI Design

UDEMY

2019

AWS Modern Applications Deploy apps reliably, rapidly, and at scale by AWS

AWS Modern Applications Deploy apps reliably, rapidly, and at scale

AWS

2020

Internet of things UI by Australian Computer Society

Internet of things UI

Australian Computer Society

2018

Certificate in Neural Networks by Australian Computer Society

Certificate in Neural Networks

Australian Computer Society

2018

Certificate in Machine Learning by Australian Computer Society

Certificate in Machine Learning

Australian Computer Society

2018

Certificate for social enterprise by Big Issue

Certificate for social enterprise

Big Issue

2019

Certificate for entrepreneurship by CERI

Certificate for entrepreneurship

CERI

2021