Licia He

Artist

Licia He · eyesofpanda.com ↗

Based

Born 1992, China. Lives and works in Tokyo

Now

Full-time artist; founder of The Generative Craft Lab, Ltd

Practice

Generative systems, robotic drawing, watercolor, code, and human-machine collaboration

Raster

raster.art/artist/licia-he ↗

In the set

Running Moon ↗ · 500 editions

Links

Full bibliography ↗

Licia He works where the procedural logic of software meets the physical behavior of pigment. Trained across studio art, computer science, and information science, she writes algorithms that organize color and form, then extends their operation through drawing robots, brushes, watercolor, paper, and other responsive materials. The resulting images are neither conventional paintings translated into code nor digital compositions merely transferred to paper. They are records of a distributed process in which artist, program, machine, tool, and material each shape the final appearance. Her description of systems, robots, and materials as the three pillars of her practice provides a concise account of this deliberately hybrid method.¹

That method emerged through an unusually integrated education. He studied studio art and computer science as an undergraduate, initially concentrating on photography and printmaking, and later pursued doctoral research in information science. Restricted access to conventional studios during graduate school helped turn her attention toward web-based generative systems. An encounter with an AxiDraw plotter then offered a practical bridge between computation and the satisfaction of making a physical object.² Her work since 2018 has developed that bridge into a sustained inquiry: what kinds of feeling become possible when an algorithm issues instructions but a brush, machine, and liquid medium remain susceptible to friction, saturation, drift, and error?

A machine that can hesitate

The plotter is often associated with precision. It follows coordinate paths, repeats instructions, and converts vectors into controlled movement. He values that precision, but her practice gains force from testing its limits. She has adapted plotting systems to carry brushes and reload them with paint, treating the robot less as an industrial printer than as a participant in a studio routine. In an early account of her process, she explained that experimentation with eyeshadow helped prompt the automatic brush-reloading mechanism that later supported her watercolor work.³ A seemingly eccentric material trial therefore became technical research, while the machine became a means of enlarging what painting could encompass.

Watercolor complicates the plotter's promise of exact reproduction. Pigment concentration changes, brushes deform, water spreads through paper fibers, and one mark may alter the surface encountered by the next. These events introduce consequential differences without dissolving the algorithmic structure. He has characterized her practice as a search between structure and fluidity, precision and error.¹ The formulation is important because it refuses a simple opposition between human expression and mechanical regularity. Expressiveness arises through their contact. Code specifies relations, the robot performs gestures, and matter interprets both.

This position also connects He to the longer history of computer-generated drawing. Early practitioners used plotters because screens and color printing were limited or unavailable, making the drawing machine a primary route from computation to visible form. The Feral File exhibition +GRAPH, curated by Casey Reas, placed He among artists renewing that historical relationship between software and plotted objects.⁴ Each participant created generative software capable of producing many drawings and selected thirty variations for physical realization. In He's contribution, custom software repeatedly directed a brush between paint and paper, with individual drawings requiring extended periods of robotic rendering.⁵ The process makes duration tangible. What appears as an immediate composition is also the residue of thousands of instructed movements.

He’s plotter does not eliminate the hand. It redistributes gesture across code, mechanism, brush, water, and paper.

Running Moon: light as a generative condition

Running Moon brought this material intelligence into an Art Blocks long-form release. Issued through Art Blocks Curated on 13 July 2022, the project consists of exactly 500 editions.⁶ Its compositions investigate an encounter between moonlight and clouds through layered fields of translucent color, segmented contours, and irregular luminous openings. Rather than illustrating a particular sky, the system constructs atmospheric relationships. Circular or partially obscured light sources appear to press against veils, bands, and cellular regions whose shifting densities suggest weather without fixing a location.

He developed the work through extensive physical sketching before translating its visual concerns into a generative system. She has said that elaborate projects such as Running Moon begin with groups of sketches, preserving observation and hand-led experimentation within the later computational process.⁷ Watercolor and stained glass informed the rendering mechanism. Their shared qualities are especially apt: both organize light through colored material, both depend upon adjacency and transparency, and both make boundaries active rather than neutral. In stained glass, a dark framework separates luminous panes. In watercolor, edges arise through pooling, overlap, and absorption. Running Moon synthesizes these effects into code while retaining their visual memory.

The edition structure allows the project's central conditions to be tested repeatedly. Across 500 outputs, the moon may dominate, recede, fragment, or become legible only through the behavior of surrounding color. Palette and geometry alter the apparent temperature and depth of the scene. Some outputs read as nocturnes, while others approach radiant abstraction. Repetition here is not duplication. It establishes a field in which viewers can compare how small procedural changes redirect mood.

Live on Art Blocks: running-moon-by-licia-he ↗

From output to encounter

Long-form generative art can make variation visible at collection scale, but He's wider practice insists that computation also has bodily and social dimensions. A plotted watercolor occupies time, requires preparation, and carries evidence of contact. Her exhibition I Have To Talk To A Plotter made this relationship explicit. Presented by GAZELL.iO in 2024, the solo project combined mixed-media plotter paintings with a live machine producing works during the exhibition.⁸ The title frames operation as conversation: instructions travel from artist to code and machine, while resistance and unforeseen behavior return information to the artist.

The works in the exhibition's A Conversation series were generated through an original Python-based system and rendered with customized extra-fine brushes and watermedia.⁸ Their production demonstrates that the digital and physical are not successive, sealed stages. Material tests can reshape software, while algorithmic structures suggest new tools and gestures. This feedback distinguishes He's work from accounts of automation centered on replacement. The machine extends the studio's range because it demands care, adjustment, and interpretation.

Her public activities likewise reposition the plotter as an accessible creative instrument. At the Victoria and Albert Museum's 2024 Digital Design Weekend, Painting with Plotters invited visitors to explore the technology behind her practice and participate in generative plotter painting.⁹ A collaborative creative-coding studio developed with Jasmine Jones similarly proposed plotters as approachable CNC machines capable of connecting programming with immediate physical artifacts.¹⁰ These educational settings make process legible without reducing the artwork to a demonstration. They show that the conceptual questions in He's practice, including authorship, control, translation, and material agency, can be encountered through making.

Emotional systems

He consistently describes algorithms as vehicles for stories and emotions rather than exercises in mathematical display. This emphasis is evident in the nocturnal atmosphere of Running Moon, but it extends across her practice. Fictional Lullaby, created for +GRAPH, addresses anxious thought at night through a system that could generate millions of compositions, from which thirty plotter-rendered paintings were produced.¹¹ The enormous space of possible images contrasts with the intimacy of sleepless experience. Computation does not distance the subject. It models the recursive, shifting patterns through which emotion occupies attention.

This emotional orientation also helps explain the importance of color. He's compositions often contain sweeping arcs, translucent cells, and clustered geometries that can suggest meteorological, botanical, or bodily processes without settling into representation. Color carries transitions between calm and intensity. The algorithm supplies a framework for those transitions, yet the resulting images resist the coldness often projected onto computational art. As one critical discussion of contemporary plotter practice observes, these artists exploit papers, pigments, inks, and hybrid media to make objects materially distinct from ordinary digital prints.⁵

The significance of He's work lies in this refusal of purity. Her programs are not immaterial concepts, her robots are not neutral executors, and her paintings are not autonomous objects detached from computation. Each work is an assembly of decisions and behaviors. Running Moon offers a particularly clear statement of that position. Across its 500 editions, a stable system continually encounters difference, producing images in which light seems structured but never immobilized. The moon becomes less a depicted object than a condition for observing how code can hold atmosphere, how variation can carry feeling, and how a machine can participate in the making of tenderness.

Bibliography

The references cited in this essay are listed below. For the complete bibliography see the Licia He links page ↗.