Original date published: June 22, 2026

What I've Worked On and What I'm Working on Now

I've been reflecting on the impact I've made in my career, the impact I would like to make in the future, and the problems that deeply interest me in this chaotic technology landscape.

I started my professional career building large-scale data pipelines to gather distributed health data for early pandemic and health outbreak detection in developing countries. This involved creating efficient data-gathering tools that could be easily deployed in local hospitals and low-resource environments. In addition, I implemented a data search and scraping tool that would gather health-related news, reports, and surveys from relevant websites to create an enriched web of information we could query alongside data obtained directly from hospitals, data aggregators, and other data vendors. This was my first attempt at leading/managing an engineering project. I was able to see how my work could drive policies and decisions that saved actual lives.

During my MSc, I had the opportunity to dive deeper into my curiosity about AI and robotics. LLMs were just taking off, and agents were still on the fringes of possibility. I decided that LLMs would be very effective for creating embodied intelligence, so I embarked on making an LLM effective at controlling a robot arm in a collaborative setting with a human. The human could issue voice commands, and the robot would execute tasks within its scene based on those commands. I continued the investigation as a research associate, exploring how foundation models could be used beyond manipulation for more complex interaction and navigation.

Following my master's, I got involved with an Innovate UK-funded project to build vertically integrated vehicle telemetry and route optimization systems for the Birmingham City Council's home-to-school transportation vendors. During the project, I was mostly forward-deployed with the largest logistics vendor, working directly with their operations and maintenance teams to build telemetry hardware and AI software pipelines for vehicle route optimization, predictive maintenance, and orchestration. The project is currently projected to deliver a 10x return on the initial investment.

Through this journey, I got to work across the entire technology stack: embedded systems, designing PCBs, circuits, and hardware components; robotics engineering, hacking ROS nodes and working on visual calibration to make trajectory execution reliable; software engineering across infrastructure deployments, backend, frontend, and mobile engineering; systems thinking; and systems engineering. I’ve learnt a lot from these experiences, but now I think mostly about what is next for my career. The world is changing rapidly with AI. There is an avalanche coming for the entire economy that will create a period of deep uncertainty and opportunity. What are the most important problems I care about now that are going to be extremely important in the future? I’ve articulated some of the problems I am excited about and will be focusing on moving forward.

I’ve spent a lot of time away thinking about what to work on next. There are three key market trends and positions I have observed over the past few months: the disaggregation of frontier labs into various independent vendors forming a robust AI ecosystem; agent security and safety becoming practical requirements, rather than buzz terms, for teams looking to unleash AI within their companies and personal lives; and physical AI infrastructure and tooling lagging significantly behind their digital counterparts.

After a few weeks of synthesis, here are some open problems I am looking to explore in the coming months:

• Domain-specific evaluation construction tooling for enterprises and AI-native companies/labs/data vendors

  Agentic evaluations currently skew towards software engineering or generalist use cases. As enterprises and AI-native companies start to deploy their own AI tooling, they are evaluating open-weights models as well as proprietary models. While frontier proprietary models lead in benchmarks, benchmark performance often does not represent an organization's deployment needs. Cheaper, less performant models may be more than enough to meet those needs, yet the bills for proprietary model use continue piling up. I want to build tools that allow such organizations to easily create/generate evaluations and explore the Pareto frontier of performance, cost, and latency for their AI deployment use cases.

• Runtime safety and security for AI agents

  AI safety is a real risk factor in AI deployment, from simple model misbehaviors to adversarial actors taking advantage of open AI systems. As companies consider moving away from API-provided frontier models to more economical open-weights models, they will be directly responsible for the security of their AI models and agents. The biggest hindrance to the wide adoption of AI agents such as OpenClaw and Hermes is the immense security vulnerabilities that these AI systems expose users to. There has been a lot of work on red-teaming AI models. I’m interested in creating blue-teaming techniques to prevent AI deployments from harming themselves or their users.

• Computer/Mobile Use Agents

  There has been an explosion of terminal-use agents. In fact, what makes agents like Claude Code and Codex very powerful is their ability to execute arbitrary terminal commands to achieve a particular outcome. While this works for most use cases, as evidenced by the broad and rapid adoption of AI agents, there is still a gap in having the agent manually manipulate the user interface. A lot of people have flagged this as a dead end and unnecessary: most modern tools have APIs, and those that don’t can simply be bypassed by writing a scraper or some type of RPA software. These are valid arguments. The goal is not just to have the agent execute tasks but to have it execute them within the same context and environment as the “user.” I believe that computer-use agents working reliably under tight time constraints, while optionally collaborating with the user, will create a new wave of opportunities in AI and downstream applications.

• Visuospatial memory systems for physical AI agents

  We have seen a rise in the use of epistemic memory for AI agents through advanced retrieval techniques such as GraphRAG. These techniques have become incredibly valuable infrastructure assets, as a good memory system is critical for agent performance in long-horizon/complex tasks. Physical AI agents are embodied in some form to perceive and, in some cases, manipulate the physical world. This means that they operate in an information space that goes beyond text. Rather, they act on vast amounts of visual data from camera sensors. Passing all of this information into the context would be practically impossible. Understanding the right video frames and time steps to pass into the context in the most efficient way possible is an incredibly challenging open problem that current AI memory providers are not positioned to solve. They have designed their systems around large amounts of non-temporal text data. This problem involves working with large streams of vision data that are both space- and time-dependent, i.e., information retrieval would be based not just on the content but also on relevant physical metadata about the data point. This would unlock many new opportunities in personal robotics, intelligent smart glasses, and AI visual inspections.

These are the core ideas I’m going to be exploring in the coming months. I’ll share my findings and expand on them as I gather more context on the problem and interact with more users and engineers tackling these challenges.

Last updated at: June 22, 2026